Free Essay

Student

In: Business and Management

Submitted By sallyjiajia
Words 12182
Pages 49
International Journal of Banking and Finance
Volume 9 | Issue 1 Article 3

6-5-2012

Modelling and forecasting volatility in the gold market
Stefan Trück
Macquarie University, stefan.trueck@mq.edu.au

Kevin Liang
Macquarie University

Follow this and additional works at: http://epublications.bond.edu.au/ijbf Recommended Citation
Trück, Stefan and Liang, Kevin (2012) "Modelling and forecasting volatility in the gold market," International Journal of Banking and Finance: Vol. 9: Iss. 1, Article 3. Available at: http://epublications.bond.edu.au/ijbf/vol9/iss1/3

This Journal Article is brought to you by the Faculty of Business at ePublications@bond. It has been accepted for inclusion in International Journal of Banking and Finance by an authorized administrator of ePublications@bond. For more information, please contact Bond University's Repository Coordinator.

Trück and Liang: Forecasting volatility in the gold market

International Journal of Banking and Finance, Volume 9 (Number 1), 2012: pages 48-80

MODELLING AND FORECASTING VOLATILITY IN THE GOLD MARKET
Stefan Trück and Kevin Liang Macquarie University, Australia

_____________________________________________
Abstract

We investigate the volatility dynamics of gold markets. While there are a number of recent studies examining volatility and Value-at-Risk (VaR) measures in financial and commodity markets, none of them focuses on the gold market. We use a large number of statistical models to model and then forecast daily volatility and VaR. Both insample and out-of-sample forecasts are evaluated using appropriate evaluation measures. For in-sample forecasting, the class of TARCH models provide the best results. For out-of-sample forecasting, the results were not that clear-cut and the order and specification of the models were found to be an important factor in determining model’s performance. VaR for traders with long and short positions were evaluated by comparing failure rates and a simple AR as well as a TARCH model perform best for the considered back-testing period. Overall, most models outperform a benchmark random walk model, while none of the considered models performed significantly better than the rest with respect to all adopted criteria. Key Words: Gold Markets, Volatility, Forecasting, Value-at-Risk, Backtesting JEL classification: G17, C22, G32

____________________________________________

1. Introduction The recent global financial crisis has highlighted the need for financial institutions to find and implement appropriate models for risk quantification. Hereby, in particular Value-at-Risk (VaR) and volatility estimates were subject to significant changes during 2007-9 financial turmoil in comparison to normal market behaviour. Further, as the risk in equity and bond markets was increasing, there was a particular interest of investors to increase their positions in the gold market. This study evaluates the

48

Produced by The Berkeley Electronic Press, 2012

1

International Journal of Banking and Finance, Vol. 9, Iss. 1 [2012], Art. 3

effectiveness of various volatility models with respect to forecasting market risk in the gold bullion market. While there is a stream of literature examining performance of models for volatility and VaR, this is a pioneer study to particularly focus on the gold market. Despite the important role gold plays for risk management and hedging in financial markets, there has been relatively little literature on the estimation of volatility of gold. Exceptions include the studies by Mills (2003), Tully and Lucey (2006), Canarella and Pollard (2008), Morales (2008) and Jun (2009). Generally, the gold market has a significant and unique role in financial markets as a safe haven that is also used for hedging and diversification. While there is no theoretical reason why gold is referred to as a safe haven asset, historical evidence suggests that investments in the gold market spikes during times of turmoil in other financial markets. One explanation could be that it is one of the oldest forms of money and was traditionally used as an inflation hedge. Moreover, gold is often uncorrelated or even negatively correlated with other types of assets. This is an important quality that allows gold to act as a diversification asset in portfolios, since a more globalised market has led to the increase in correlation among other assets. This became also evident during the financial crisis of 2007-2009 where the negative effect of one market readily flowed into other markets, yet the gold market remained relatively unscathed during this period of turbulence. So far there has been no study using volatility and VaR modelling in the spot gold and gold futures markets. Gold market research has concentrated on the role of gold as a hedging or diversification tool, in particular as a safe haven during market crashes. This study examines various models that can be used in forecasting volatility, to evaluate their respective performance. Finding appropriate models for volatility is of interest for several reasons: firstly, it is an integral factor of derivative security pricing, for example, in the classic Black-Scholes model or alternative option pricing formulas. Secondly, as a representation of risk, volatility plays an important role in an investor’s decision making process. Volatility is not only of great concern for investors but also policy makers and regulators who are interested in the effect of volatility on the stability of financial markets in particular and the whole economy in general. Finally, volatility estimation is an essential input in many VaR models, as well as for a number of applications in a firms market risk management practices.

http://epublications.bond.edu.au/ijbf/vol9/iss1/3

2

Trück and Liang: Forecasting volatility in the gold market

The remainder of the paper is set up as follows. Section 2 provides a brief review on the global gold market and studies on volatility modelling of financial markets in general and gold markets in particular. Section 3 provides an overview on the data and techniques used in this study. In particular various models for volatility forecasting and evaluating model performance are reviewed. Empirical results of the study are reported in Section 4 while section 5 concludes.

2. Gold Markets and Volatility Models 2.1 The gold market Gold has been used throughout history as a form of payment and has been a standard for currency equivalents to many economic regions or countries. In spite of its historical monetary significance, a free functioning world market only came of age in recent times. Before 1971, the gold standard was mostly used in various times in history, where domestic currencies have been backed by gold. The system existed until 1971, when the US stopped the direct convertibility of the United States dollar to gold, effectively causing the system to break down. Since then, a global market for gold in its own right developed, remaining open around the clock and open to a range of derivative instruments. The market for gold consists of a physical market in which gold bullions and coins are bought and sold and there is a paper gold market, which involves trading in claims to physical stock rather than the stock themselves. Physical gold is generally traded in the form of bullions. The bullion market serves as a conduit between larger gold suppliers such as producers, refiners and central banks and smaller investors and fabricators. The bullion market is essentially a spot market, but is complemented by the use of forward trading for the hedging of physical positions. Since 1919, the most widely accepted benchmark for the price of gold is known as the London gold fixing, a twice-daily (telephone) meeting of representatives from

Produced by The Berkeley Electronic Press, 2012

3

International Journal of Banking and Finance, Vol. 9, Iss. 1 [2012], Art. 3

five bullion-trading firms. 1 Furthermore, there is active gold trading based on the intra-day spot price, derived from gold-trading markets around the world as they open and close throughout the day. The key prices in the London bullion market are the spot (fixings) price, the forward price and the lease rate. The spot (fixings) price is a daily clearing or fix price obtained by balancing purchases and sales ordered through its members. The forward price (GOFO) is the simultaneous purchase and sales price of gold forward contracts of various lengths. Generally, the GOFO rate is expressed as an annual percentage. Finally, the lease rate refers to short-term loans denominated in gold and is expressed as an annualized interest rate. Since 1971 the price of gold has been highly volatile, ranging from a high of US$850 on January 21, 1980, to a low of US$252.90 on June 21, 1999. The period from 1999 to 2001 marked the so-called Brown Bottom after a 20-year bear market. Prices increased rapidly from 1991, but the 1980 high was not exceeded until 2008 when a new maximum of $865.35 was set on January 3, 2008. Another record price was set on March 17, 2008 at $1023.50. In the second half of 2009, gold markets experience renewed momentum upwards due to increased demand and a weakening US dollar. Overall, since April 2001, the gold price has more than tripled in value against the US dollar. 2.2 Factors influencing gold prices As mentioned above, gold has a unique place in financial markets. Of all the precious metals, gold is the most popular as an investment. Investors generally buy gold as a hedge or safe haven against any economic, political, social or currency-based crises. These crises include investment market declines, burgeoning national debt, currency failure, inflation but also scenarios like war or social unrest. As in any commodities, the price of gold is ultimately driven by its supply and demand. However, unlike other resources, hoarding and disposal plays a much bigger role in price formation because most of the gold ever mined still exists and is potentially able to enter the market for the right price. Given the huge quantity of stored gold,

1

All five members - Bank of Nova Scotia–ScotiaMocatta, Barclays Bank Plc, Deutsche Bank AG, HSBC Bank USA, NA and Société Générale are market making members of the London Bullion Market Association (LBMA).

http://epublications.bond.edu.au/ijbf/vol9/iss1/3

4

Trück and Liang: Forecasting volatility in the gold market

compared to the annual production, the price of gold is mainly affected by changes in sentiment, rather than changes in the actual annual production. Also macroeconomic factors such as low real interest rates can have an effect on gold price. If the return on bonds, equities and real estate is not adequately compensating for risk and inflation, then the demand for gold and other alternative investments such as commodities increases. An example of this is the period of stagflation that occurred during the 1970s which led to an economic bubble forming in precious metals. Financial market declines such as the 2007-9 global financial crisis usually leads investors to look for alternative and less volatile investment opportunities for their funds. It will also increase the need for investors to hedge their portfolios to minimise their risk in case of further decline. The demand for gold and, thus, its price increase, empirically is due to the role of gold as a safe haven in times of crises. This is one of the major reasons to drive gold prices to new highs throughout the post-financial crisis period. Central banks and the International Monetary Fund (IMF) also play an important role in determining the gold price. At the end of 2004 central banks and official organizations held 19 percent of all above-ground gold as official gold reserves. Thus, they have a significant influence on the gold market not only as a major buyer and seller. Also, speculation on their future gold holding levels can also be a driving factor. Recently, the assumption that central banks around the world will increase their gold reserve levels as a hedge against the falling US dollar has also contributed to the rise of gold prices. The performance of gold bullion is often compared to stocks. However, they are fundamentally different asset classes. Gold is regarded by some as a store of value (without growth) whereas stocks are regarded as a return on value. Stocks and bonds perform best in periods of economic stability and growth, whereas gold is seen as the asset to hold in times of uncertainty and crisis. Throughout history there has been a cyclical run with long periods of stock outperformance followed by long periods of gold outperformance. Over the long term, equity markets have been able to outperform gold overall.

Produced by The Berkeley Electronic Press, 2012

5

International Journal of Banking and Finance, Vol. 9, Iss. 1 [2012], Art. 3

2.3 Volatility Models Within the last three decades various approaches to volatility modelling have been suggested in the econometric and financial literature. In the following we will provide a brief overview of developments in the literature starting with the autoregressive conditional heteroskedasticity (ARCH) models (Engle, 1982). Bollerslev (1986) introduced the generalised ARCH (GARCH) model. The latter is often utilised in financial market studies. The general idea is to predict the current period’s variance by forming a weighted average of a long term average, the forecasted variance from last period, and information about volatility observed in the previous period. If the return is unexpectedly large either in the upward or the downward direction, then the trader will increase the estimate of the variance for the next period. This model is also consistent with the volatility clustering often seen in financial returns data, where large changes in returns are likely to be followed by further large changes. Since the introduction of these models, they have been widely used in volatility modelling and forecasting. Researchers such as French et al. (1987) and Akgiray (1989) utilised GARCH models to capture the behaviour of stock market price volatilities. Argiray (1989) compared the GARCH (1,1) model to other historical estimation methods and found that the GARCH (1,1) model outperformed its competitors. Many extensions of the GARCH model have been introduced in the literature since: e.g. GARCH-in-mean (GARCH-M) models (Engle et al., 1987), EGARCH models (Nelson, 1991), Threshold ARCH (TARCH) and Threshold GARCH (TGARCH) (Glosten, Jaganathan, and Runkle, 1993; Zakoïan, 1994) and Power Arch (PARCH) models (Ding et al., 1993) just to name a few. A number of studies have focused on optimal model specification and the performance of various GARCH models in financial markets providing no clear-cut results. Hansen and Lunde (2005) carried out comprehensive testing of 330 variants of ARCH type models on their performance in estimating volatility in exchange rates and stock returns. The study found that the GARCH (1,1) model outperforms other models in estimating exchange rate volatilities but underperforms in estimating stock returns. McMillan et al. (2000) tested a set of ten volatility estimation models including

http://epublications.bond.edu.au/ijbf/vol9/iss1/3

6

Trück and Liang: Forecasting volatility in the gold market

random walk, moving average and GARCH models in forecasting UK stock market returns at different frequencies. They found that the performance of each model varied depending on the length of frequencies, the series as well as the type of loss function being applied. The random walk model outperformed others at the monthly frequency, while GARCH and moving average models were superior using daily forecasts. Brooks and Persand (2002, 2003) examine various ARCH and GARCH type models with respect to volatility forecasting. They report that, while the forecasting performance of the models depended on the considered data series and time horizon, the overall most preferred model is a simple GARCH(1,1). This is also consistent with many other studies such as e.g. Bollerslev et al. (1992). On the other hand, Braisfold and Faff (1996) evaluate volatility models in forecasting stock returns, and find that none of the models significantly outperforms the others. Recently, also a stream of literature has emerged focusing on modelling and forecasting volatility with respect to the quantification of Value-at-Risk (VaR). As pointed out by Jorion (1996), VaR plays a substantial role in managing risks for financial institutions. The importance of the VaR measure is further highlighted by regulators in the Basel Committee on Banking Supervision. 2 The performance of volatility models with respect to appropriate quantification of VaR has been investigated by Danielsson and De Vries (2000): conditional parametric methods such as the GARCH model significantly underpredict the VaR of U.S. stock returns. Giot and Laurent (2001, 2003) investigate volatility models for both negative and positive returns, with the latter representing risk for short position holders. They find that skewed asymmetric ARCH models using the Student t distribution perform best with respect to risk quantification. Sadorsky (2006), investigating oil price volatility, tested a great variety of volatility models by evaluating the forecasting performance using different VaR measures. His findings suggest that while no model could consistently outperform the others, a GARCH model as well as a TGARCH performed quite well for modelling and forecasting the volatility and risk of oil prices. Tully and Lucey (2006) examine various macroeconomic influences on gold using models including the asymmetric power GARCH model (APGARCH) for spot
The Basle Committee on Banking Supervision announced in April 1995 that capital adequacy requirements for commercial banks are to be based on VAR.
2

Produced by The Berkeley Electronic Press, 2012

7

International Journal of Banking and Finance, Vol. 9, Iss. 1 [2012], Art. 3

and futures prices over a 20 year period, paying special attention to periods of stock market crashes. Their results suggest that the price of gold is significantly influenced by the U.S. dollar while during periods of financial crises an APGARCH model performs best with respect to volatility. Mills (2003) investigates the statistical behaviour of daily gold prices, and finds that price volatility scaling with long-run correlations is important while gold returns are characterised by short-run persistence and scaling with a break point of 15 days. Canarella and Pollard (2008) apply power GARCH model to the London Gold Market Fixings to investigate long memory features as well as conditional volatility behaviour of the returns. They find that APGARCH models were able to adequately capture long memory in returns and that market shocks have strong asymmetric effects: conditional volatilities of gold prices are affected more by good news (positive shocks) than bad news (negative shocks). Morales (2008) discusses volatility spill-over effects between precious metal markets using GARCH and EGARCH techniques. Gold was found to be influenced by prices of other precious metals, but there was little evidence to suggest other precious metals influencing gold prices.

3. Data and Models 3.1 The Data The data for this study are daily PM gold fixing prices on the London Bullion Market available from the official The London Bullion Market Association website (www.lbma.org.uk). The market is a wholesale over-the-counter (OTC) market for gold and silver. The fixings are the internationally published benchmarks for precious metals. The Gold Fixing is conducted twice a day by five Gold Fixing members, at 10:30 am and 3:00 pm. This study will use the daily PM fixings price released at 3:00 pm as quoted in USD. The data cover 2508 observations from 4 January 1999 to 30 December 2008. The time series exhibits a number of price shocks, e.g. during the periods around September 11, 2001, the beginning of the Iraq war in 2003 as well as the global financial crisis in 2008.

http://epublications.bond.edu.au/ijbf/vol9/iss1/3

8

Trück and Liang: Forecasting volatility in the gold market

For the observed gold fixing prices pt, the daily log-returns are calculated as rt = ln (pt/pt-1). Table 1 provides a summary of descriptive statistics for the considered return series. Table 1: Summery Statistics of Gold-Fixing Log Returns Gold Fixing Mean Median Maximum Minimum Std. Dev. Skewness Kurtosis Jarque-Bera Probability Sum Sum Sq. Dev. Observations 0.044204 0.045767 7.005954 -7.971887 1.143359 -0.053361 8.533989 3200.230 0.000000 110.8201 3276.017 N = 2507

We observe that the mean and median of daily returns are positive indicating that overall gold prices were increasing during the considered time period. The magnitude of the average return (0.044%) is very small in comparison to its standard deviation (1.14%). Further, the large kurtosis of 8.53 indicates the leptokurtic characteristics of daily returns. Figure 1: Time series and distribution of returns for gold fixing prices, 1999 to 2008
R_DAILY
8
700 600 500
Frequency

R_DAILY

4

0

400 300 200

-4

-8
100

-12 500 1000 1500 2000 2500

0 -8 -6 -4 -2 0 2 4 6 8

Produced by The Berkeley Electronic Press, 2012

9

International Journal of Banking and Finance, Vol. 9, Iss. 1 [2012], Art. 3

Obviously, the series has a distribution with tails that are significantly fatter than those of a normal distribution. This indication of non-normality is also supported by the Jarque and Bera (1980) test statistic, which rejects the null hypothesis of a normal distribution at all levels of significance. Figure 1 provides a plot of the time series for the daily log-returns as well as a histogram of the return distribution. The figures indicate heteroscedasticity and volatility clustering for the return series that also exhibits a number of rather isolated extreme returns caused by unforeseen events or shocks to the gold market. We further test for stationarity of the return series using the Augmented Dick Fuller (1979) (ADF) and Phillips Perron (1988) (PP) unit root tests. The ADF test is set to a lag length 0 using the Schwarz Information Criterion (SIC) and the PP test is conducted using the Bartlett Kernel spectral estimation method. Results are reported in Table 2, and indicate that for both tests the null hypothesis of a unit root is rejected. So the return series gold fixing prices can be considered to be stationary. Table 2: Results for Augmented Dickey-Fuller and Phillips-Perron unit root tests for gold fixing return series (4 January 1999 -30 December 2008) t-Statistic -49.24794 Adj. t-Stat -49.29521 Prob.* 0.0001 Prob.* 0.0001

Augmented Dickey-Fuller test statistic

Phillips-Perron test statistic

3.2 Considered Models In the following, a variety of models is introduced for volatility modelling and forecasting of the daily returns. We will follow several studies in the literature, see e.g. Sadorsky (2006), and measure the volatility of gold by its squared daily return: ����2 = ����2 �
(1)

Thus, most of the models will be evaluated with respect to their ability to model and forecast the volatility measured by the squared return of the gold fixings price. The first model to be considered in the empirical analysis is a random walk model

http://epublications.bond.edu.au/ijbf/vol9/iss1/3

10

Trück and Liang: Forecasting volatility in the gold market

(RW). If the volatility of gold market returns follows a random walk, the best forecast for the next period’s volatility is the volatility observed in the current period: ����+1 = ����2 �2
(2)

This random walk model will be used as a benchmark model for the out-of-sample performance of the estimated models. The second standard class of models to be considered are historical mean (HM) models. In these models, the forecast for the volatility of the next period is the average uncorrelated with other observable variables and if ����2 is uncorrelated with its own � Defining ����2 = ����2 , the HM model can be denoted by
2 ����+1 = �� ���=0 ����−�� �2 1 ��−1

of all previous volatilities. In particular, if ����2 is a random variable, which is �

past values, then the population mean can be considered as the optimal forecast.

(3)

A popular alternative to the HM model is the m-period moving average (MA) model. The forecast for the next period is based on the average of the last m observations. A value for m has to be determined. We decided to use moving averages of length m=20, 40 and 120 days, corresponding to about one month, two months and six months. The MA(m) model can be denoted by:.
2 ����+1 = �� ���=0 ����−�� �2 1 ��−1

(4)

The next model we consider is the exponentially weighted moving average model (EWMA). It forecasts the future volatility by applying weighting factors which decrease exponentially. That is, the method gives higher weights to more recent 1) and the value of volatility ����2 observed on day t: ����+1 = ������2 + (1 − ��)����2 �2 � the weighted average of the estimated volatility ����2 for day t (made at the end of day t� observations while still not discarding older observations entirely. It is calculated as
(5)

The smoothing parameter α governs how responsive the forecast is to the most recent daily percentage change. Generally, α lies between 0 and 1, and the process becomes a RW for α =0. A popular choice for the parameter α is based on J.P.

Produced by The Berkeley Electronic Press, 2012

11

International Journal of Banking and Finance, Vol. 9, Iss. 1 [2012], Art. 3

Morgan’s RiskMetrics (1995) where it is suggested that α = 0.94 provides forecasts of the variance rate closest to the actual variance rate for a range of different market variables. An alternative is an ordinary least squares (OLS) model. The relationship between volatility on day t and day t+1 is described based on a linear relationship: ̂ ̂ � ����+1 = ��0 + ��1 ����2 �2
(6)

The parameter estimates are then determined by OLS estimation. The model can be extended to an autoregressive (AR) model of order p where the current volatility is a linear function of the last p observations for the volatility. We implement a model of order p = 5 such that we estimate an AR(5) model that can be described by the following equation: ̂ ̂ � ̂ �2 ̂ �2 ̂ �2 ̂ �2 ����+1 = ��0 + ��1 ����2 + ��2 ����−1 + ��3 ����−2 + ��4 ����−3 + ��5 ����−4 �2

(7)

We also consider a weighted moving average of disturbance terms model (MAD) where the volatility in period t+1 is modelled as a function of the lagged values of the disturbance term εt. Similar to the AR model, we decided to use a MAD model of order 5 that can be described by the following equation: ����+1 = ��0 + ��1 ε�� + ��2 ε��−1 + ��3 ε��−2 + ��4 ε��−3 + ��5 ε��−4 �2 � � � � � �

(8)

We decided to also use an autoregressive moving average (ARMA) or Box-Jenkins model that includes both an autoregressive (AR) and a moving average (MAD) component. A simple ARMA(1,1) can then be described by the following equation: ̂ ̂ � ����+1 = ��0 + ��1 ����2 − ��1 ε�� �2 �
(9)

Since the introduction of autoregressive conditional heteroscedasticity (ARCH) models by Engle (1982), the ARCH and even more the related GARCH (Bollerslev, 1986) model have become standard tools for examining the volatility of financial variables. The model has proven to be very useful in capturing heteroskedastic behaviour or volatility clustering without the requirement of higher order models in various financial markets, see e.g. Choudhy (1996) or Sadorsky (2006). In a GARCH (1,1) model the conditional variance equation can be denoted by

http://epublications.bond.edu.au/ijbf/vol9/iss1/3

12

Trück and Liang: Forecasting volatility in the gold market

while the equation for the conditional mean is

such that the one day ahead variance forecast can be expressed as: � ̂� ℎ��+1 = �� + �� �� + �� ℎ�� � ���̂ 2
(12)

���� = �� + ���� , ���� ~��(��, ℎ�� )

2 ℎ�� = �� + ������−1 + ��ℎ��−1

(10)

(11)

A popular extension of the GARCH (1,1) model is also the GARCH in mean (GARCH-M) model that was first proposed by Engle et al. (1987). The GARCH-M model includes the conditional variance in the specified equation for the conditional mean. This allows for so-called time varying risk premiums. Chou (1988) suggests that the dynamic structure of the conditional variance can be captured more flexibly by a GARCH-M model, using the following specification for the conditional mean: ���� = �� + ��ℎ�� + ���� , ���� ~��(��, ℎ�� )
(13)

Another extension of standard ARCH and GARCH models has been suggested by Glosten et al. (1994) and Hentschel (1994): threshold ARCH (TARCH) and GARCH (TGARCH) models, which are popular in describing return asymmetry. Large negative returns are often followed by a substantial increase in volatility such that the TARCH and TGARCH models distinguish between negative and positive returns. The TGARCH model that will be considered in the empirical analysis treats the conditional standard deviation as a linear function of shocks and lagged standard deviations (Hentschel, 1994) and is denoted by:
2 2 ℎ�� = �� + ������−1 + ��ℎ��−1 + ������−1 ����−1

where ����−1 is equal to 1 if εt < 0, and zero otherwise. Obviously, in this model,
2 2 ����−1 > 0, and ����−1 < 0 will have different effects on the conditional variance. If

(14)

increase volatility and there is evidence of a leverage effect.

�� ≠ 0, there is asymmetry in the model. If �� > 0, the occurrence of bad news will

Produced by The Berkeley Electronic Press, 2012

13

International Journal of Banking and Finance, Vol. 9, Iss. 1 [2012], Art. 3

3.3 Performance Evaluation Measures To evaluate the performance of the considered models, we apply a variety of measures such as mean squared error (MSE), root mean squared error (RMSE), mean absolute deviation (MAD), mean absolute percentage error (MAPE) and the Theil U statistic. The MSE quantifies the difference between predicted and actually observed values by considering the squared difference between these two quantities:

= MSE

1 T ∑ (σ t2 − σˆ t2 )2 t t =1

(15)

The RMSE is simply the root of MSE and has the advantage of being measured in the same unit as the forecasted variable:

= RMSE

1 T ∑ (σ t2 − σˆ t2 )2 t t =1

(16)

The MAE is also measured in the same unit as the forecast, but gives less weight to large forecast errors than the MSE and RMSE:

= MAE

1 T ∑ | σ t2 − σˆ t2 | t t =1

(17)

The MAPE measures the forecast quality independent of the unit of measurement of the variable. The measure might be less useful when the actual values of σ t2 are close to zero, because in this case the MAPE will take on large values even if the errors are fairly small in magnitude. Another drawback of the MAPE is that if there are zero values (which may happen for daily squared returns) there will be a division by zero. We still decided to examine the results using the MAPE measure that can be denoted by:
2 ˆ2 100 T | σ t − σ t | MAPE = ∑ σ t2 t t =1

(18)

We also investigate the forecasting performance using the Theil U statistic that examines the RMSE measure of a forecast against a naïve one step ahead forecast. If

http://epublications.bond.edu.au/ijbf/vol9/iss1/3

14

Trück and Liang: Forecasting volatility in the gold market

the Theil U statistic is smaller than 1, the tested forecast model outperforms the naïve model: if the U statistic is larger than 1, the naïve forecast is the better model. Note that in our analysis we decided to use the RW model as the naïve benchmark model for forecasting. �� =
��������(����������������)

��������(��������������������������)

(19)

While the above forecasting quality measures are useful for providing different performance measures on applied models, they do not statistically test if the models are significantly different or better from another. Therefore, we will also apply the Diebold-Mariano (1995) test (DM) to compare the predictive ability between two forecasting models. The null hypothesis of the test is that the predictive ability of two forecasting models is the same. In our empirical analysis, we are particularly interested whether our forecast models are able to significantly outperform a simple RW model such that the considered models are tested against the RW model using a simple t-test, see e.g. Diebold (1998). Thus, the null hypothesis of equal performance of the models is rejected when the test-statistic �� = ��⁄ �
1 � �� √��

(20)

with

yields significant values. In the empirical analysis we will restrict ourselves to oneperiod-ahead forecasts only. Note that the test could also be applied to k-step-ahead forecasts, see e.g. Diebold and Mariano (1995). The authors point out that the test tends to be less accurate for small sample sizes and k-step-ahead forecasts. However, these issues are unlikely to affect our empirical analysis due to a comparably large sample size and the use of one-period-ahead forecasts only.

2 2 �� = �� ∑�� (��1�� − ��2�� ) ��=1

(21)

Produced by The Berkeley Electronic Press, 2012

15

International Journal of Banking and Finance, Vol. 9, Iss. 1 [2012], Art. 3

4. Empirical Results 4.1. In-sample forecasting performance In this section, we compute the one-step-ahead volatility forecasts using the models described in the previous section. For the in-sample analysis, the data are divided into three sub-periods: sub-period 1 from 28th Jun 1999 - Dec 2004, which is a period of slightly increasing gold prices over 6 years; sub-period 2 from Jan 2005 - Dec 2007, which is a period of substantially increasing gold prices over 3 years; and sub-period 3 from Jan 2008 - Dec 2008, a period of very volatile gold prices. For in-sample forecasting, all observations within the period are used to estimate the models, and the forecasting results are compared to the actual values. The complete results including the ranking for each measure in each sub-period are presented in tables 3, 4 and 5, respectively. Obviously, the forecast error statistics, MSE, RMSE and MAE depend on the scale of the dependent variable and the differences between the actual and forecasted volatility. On the other hand, the MAPE does not depend on the scale of the variable: given a perfect fit of a model, the MAPE would be zero while there is no upper bound for the MAPE. For all measures, it can be concluded that the smaller the error statistic, the better the forecasting ability of a model. In order to facilitate the comparison of the models, for each performance criteria also the relative ranks are provided. In the first sub-period, the price fluctuations were relatively low with a general upward trend. Only one structural break occurred after the 11 September 2001 attack. The shock lasted only a short period of time and did not have long-lasting effects on the volatility. The results show that for the MSE, MAE, RMSE and Theil U criteria the TARCH model yields the best results with respect to in-sample one period ahead forecasts. The model further ranks second for the MAPE criterion. Overall, most test statistics are consistent in ranking the forecasting performance of the considered models, with the exception of the MAPE measure. However, as discussed earlier, the MAPE can be unreliable in cases where the denominator contains the value of zero as it was the case in our evaluation where results were removed when the daily change and volatility were zero.

http://epublications.bond.edu.au/ijbf/vol9/iss1/3

16

Trück and Liang: Forecasting volatility in the gold market

Table 3: In-sample Results for Sub-period 1

MSE Rank RMSE Rank MAE Rank MAPE Rank Theil U Rank DM Rank

RW 10.231 13 3.199 13 1.209 13 8264.1 1 1.000 13 t-stat

HM 7.617 10 2.760 10 1.054 6 11071.5 13
0.863

MA(20)

MA(40)

MA(120)

9
-1.217

7.937 12 2.817 12 1.064 8 8413.4 3 0.881 12
-1.082

0.140 p-value Table 3 In-sample forecast results for sub-period 1 (June 1999 - December 2004) for examined models and the considered performance evaluation measures MSE, RMSE, MAE, MAPE, Theil U as well as results for Diebold-Mariano test with RW model as benchmark.

9 0.112

12

7.798 11 2.793 11 1.058 7 10260.2 9 0.873 10 -1.140 10 0.127

7.615 9 2.760 9 1.066 9 10723.3 12 0.863 8 -1.228 8 0.110

OLS 6.774 4 2.603 4 1.029 4 10263.7 10 0.814 4 -1.893 4 0.029

AR(5) 6.708 2 2.590 2 1.014 2 9707.6 5 0.810 2 -1.912 2 0.028

MAD(5) 6.718 3 2.592 3 1.018 3 9913.2 8 0.810 3 -1.903 3 0.029

ARMA(1,1) 7.615 8 2.760 8 1.065 9 10722.3 11 0.863 7 -1.228 7 0.110

EWMA 7.472 6 2.733 6 1.032 5 9114.0 4 0.855 6 -1.313 6 0.095

GARCH(1,1) 7.907 11 2.812 11 1.100 12 9892.5 6 0.879 11 -1.082 11 0.140

GARCH-M 7.258 5 2.694 5 1.064 8 9904.0 7 0.842 5 -1.457 5 0.073

TARCH 5.713 1 2.390 1 0.990 1 8288.1 2 0.747 1 -2.348 1 0.010

48

Produced by The Berkeley Electronic Press, 2012

17

International Journal of Banking and Finance, Vol. 9, Iss. 1 [2012], Art. 3

Table 4: In-sample Results for Sub-period 2
RW 15.203 13 3.899 13 1.785 13 14790.9 4 1.000 13 t-stat p-value HM 8.347 12 2.889 12 1.325 4 16439.6 8
0.741

MA(20)

MA(40)

MA(120)

MSE Rank RMSE Rank MAE Rank MAPE Rank Theil U Rank DM Rank

12
-2.819

7.704 5 2.776 5 1.426 9 15475.7 5 0.712 5
-3.113

0.002 Table 4 In-sample forecast results for sub-period 2 (January 2005 - December 2007) for examined models and the considered performance evaluation measures MSE, RMSE, MAE, MAPE, Theil U as well as results for Diebold-Mariano test with RW model as benchmark.

12 0.002

6

7.706 6 2.776 6 1.422 7 15515.4 6 0.712 6 -3.093 7 0.002

7.829 8 2.798 8 1.424 8 19616.2 11 0.718 8 -3.038 9 0.002

OLS 8.153 9 2.855 9 1.487 12 22234.2 13 0.732 9 -2.932 11 0.002

AR(5) 7.595 3 2.756 3 1.467 10 18682.8 10 0.707 3 -3.055 8 0.002

MAD(5) 7.762 7 2.786 7 1.481 11 19663.1 12 0.715 7 -3.033 10 0.002

ARMA(1,1) 7.558 2 2.749 2 1.418 5 17411.6 9 0.705 2 -3.173 3 0.002

EWMA 7.607 4 2.758 4 1.421 6 16083.5 7 0.707 4 -3.162 4 0.001

GARCH(1,1) 7.558 2 2.748 2 1.434 8 17535.2 8 0.705 2 -3.138 5 0.002

GARCH-M 7.682 6 2.772 6 1.443 9 17679.1 9 0.711 6 -3.184 2 0.001

TARCH 6.624 1 2.574 1 1.314 1 9491.3 1 0.660 1 -3.567 1 0.000

http://epublications.bond.edu.au/ijbf/vol9/iss1/3

18

Trück and Liang: Forecasting volatility in the gold market

Table 5: In-sample Results for Sub-period 3

MSE Rank RMSE Rank MAE Rank MAPE Rank Theil U Rank DM Rank

RW 102.833 13 10.141 13 5.758 13 11637.6 10 1.000 13 t-value

HM 61.154 12 7.820 12 4.668 12 6670.9 1
0.771

MA(20)

MA(40)

MA(120)

12
-9.769

50.452 8 7.103 8 4.205 8 11419.8 9 0.700 8
-12.394

0.000 p-value Table 5 In-sample forecast results for sub-period 3 (January 2008 - December 2008) for examined models and the considered performance evaluation measures MSE, RMSE, MAE, MAPE, Theil U as well as results for Diebold-Mariano test with RW model as benchmark.

12 0.000

8

49.695 4 7.049 4 4.060 4 10527.5 5 0.695 4 -12.498 6 0.000

51.953 10 7.208 10 3.820 3 8212.6 2 0.711 10 -11.948 10 0.000

OLS 53.201 11 7.294 11 4.509 11 13501.3 13 0.719 11 -11.765 11 0.000

AR(5) 48.378 3 6.955 3 4.360 9 12257.8 12 0.686 3 -12.467 5 0.000

MAD(5) 48.168 2 6.940 2 4.410 10 12197.5 11 0.684 2 -12.699 2 0.000

ARMA(1,1) 49.789 5 7.056 5 4.083 5 10654.3 6 0.696 5 -12.549 3 0.000

EWMA 50.017 7 7.072 7 4.129 6 10654.9 7 0.697 7 -12.534 4 0.000

GARCH(1,1) 49.822 6 7.058 6 4.016 4 9977.4 4 0.696 6 -12.391 9 0.000

GARCH-M 51.133 9 7.151 9 4.148 7 9912.7 3 0.705 9 -12.475 7 0.000

TARCH 44.288 1 6.655 1 3.812 2 10973.9 8 0.656 1 -13.876 1 0.000

Produced by The Berkeley Electronic Press, 2012

19

International Journal of Banking and Finance, Vol. 9, Iss. 1 [2012], Art. 3

The different ranking according to the MAPE criterion is also indicated by the fact that while the benchmark RW model ranks last for all other statistics, it is the preferred model according to the MAPE. The performances of the other GARCH type models vary substantially depending on the considered evaluation criterion. Interestingly, the GARCH (1,1) model performs poorly and is ranked in the bottom quartile of all tested models. The GARCH-M model performs better, ranking 5th out of the 13 models in most statistics. The simple AR(5) and MAD(5) models also perform well, producing the second and third lowest values for the considered error statistics. Other ARMA type models including the OLS and ARMA(1,1) rank in the middle, while models based on past averages such as the HM, and the various MA models do not perform very well. However, they still outperform the RW model for most of the considered criteria. The Theil U statistics are less than 1 for all considered models being indicative of the models performing better than the random walk model. On the other hand, the results for the DM test are not so clear-cut. Only the OLS, AR, MAD and TARCH models are significantly better than the benchmark random walk model at the 0.05 significance level. So during the period with low volatility from June 1999 and December 2004, all models seem to provide better in-sample forecasts than the RW model, while only 4 out of 12 significantly outperform the RW model according to the DM test. We will see that during the more volatile periods, the DM test provides more significant results. We could conclude that during periods of low volatility in the gold market, it seems difficult to significantly outperform the RW model using a parametric model. Overall, while the TARCH model had the lowest forecast errors of all models, one may also consider AR, MA and OLS type models when forecasting in periods of low volatility. As mentioned above, during the second sub-period from January 2005 and Dcember 2007, the gold market experienced a major boom, with gold almost doubling in value. Overall, the forecast errors for this period are slightly higher than that of subperiod 1. Once again the TARCH model turns out to be the best model considering the different performance criteria. Also the ARMA(1,1) and GARCH(1,1) models provide good results and they rank among the best three models for several of the performance criteria. Interestingly, for this period, different performance measures provide quite

http://epublications.bond.edu.au/ijbf/vol9/iss1/3

20

Trück and Liang: Forecasting volatility in the gold market

different rankings: Theil’s U allocates more weight to large deviations while MAE and MAPE provide a different weighting of the performance errors. Thus, the conditional variance type models GARCH(1,1) and GARCH-M models were selected by MSE, RMSE, and Theil’s U as the second and sixth best models. In fact, these perform worse against the absolute error measures of MAE and MAPE. The MSE and Theil’s U also indicate that the EWMA, AR(5) and ARMA models perform well. The RW is once again the worst performing model, ranking last for all statistics except MAPE. The DM values for this period are all highly significant even at the 0.01 level, indicating that most models are able to significantly outperform the RW benchmark in this period. This is also confirmed by U statistic where all models yield lower values than in the first sub-period. The U values range from 0.26 to 0.34 indicating that even the worst performing model (HM) is still significantly better than the RW benchmark. Overall, the results for the second sub-period suggest that predictive models with conditional volatility like TARCH, GARCH and GARCH-M seem to perform quite well during this period of significant increases in the gold price. The third sub-period from January to December 2008 also includes the advent of the global financial crisis, when various financial markets as well as the gold market exhibited a long period of extreme volatility. Generally, one would expect this period being the most difficult for volatility prediction. This is confirmed by both MSE and MAE-based criteria yielding clearly higher values than for the previous two subperiods. For example, the MSE is five times higher than during the first and second sub-period while the MAE increases by roughly 200 percent. Also for the third subperiod, MSE, RMSE and U favour the TARCH model as yielding the best predictions, while the AR(5) and MAD(5) rank second and third. For these criteria, the random walk model is the worst performing model, followed by the HM model. Also the MAE measure gives indication of superiority of the TARCH model over the others. However, for this criterion, the AR and MAD models perform rather poorly and only rank ninth and tenth. Again, the two worst performing models are the RW and HM model. The DM test show that for the third sub-period all models were able to significantly outperform the RW model at the 0.01 level. Results for Theil’s U are

Produced by The Berkeley Electronic Press, 2012

21

International Journal of Banking and Finance, Vol. 9, Iss. 1 [2012], Art. 3

similar to the second sub-period indicating that the models provide substantially smaller RMSE than the RW model for the volatile third sub-period. Overall, we conclude that for in-sample fit, the TARCH model can be considered as the most appropriate, ranking first for almost all of the examined performance measures and sub-periods. 4.2. Out-of-sample forecasting results In the following we report the results for an out-of-sample analysis of the models by comparing one-step-ahead volatility for the most volatile period from July 1, 2008 to December 30, 2008. A recursive window approach is used. For the recursive window approach, the initial estimation date is fixed and the models are estimated using all observations available up to the initial estimation date. It is an iterative procedure, where in each time step, the estimation sample is augmented to include one additional observation in order to re-estimate the volatility forecast for the next day. Again, results are benchmarked against a RW model. Note that despite its simplicity, particularly in out-of-sample forecasting the random walk model is often considered as a benchmark model that is difficult to beat: for example, Stock and Watson (1998) examine various US macroeconomic time series and suggest the RW model to perform best amongst a number of competing models. The out-of-sample results for the different models are provided in Table 6. Our results for the MSE criterion suggest that the MA(40) model provides the most accurate forecasts while the EWMA model ranks seconds. Interestingly, similar to the considered in-sample periods, the RW model proved to be the worst amongst the examined models also for out-of-sample forecasting. It ranked last with respect to the MSE criterion and provided predictions significantly less accurate than most of the considered models. Another feature of the results is that there are only relatively small differences with respect to MSE among the ten best models: the MSE for the MA(40) model is 83.29 while the MAD(5) model provides a MSE of 90.08. With respect to MAE, we observe the smallest error for the MA(120) model. The HM and MAD models, also perform well, ranking second and third, respectively. The benchmark RW model is substantially less accurate than the other models. Again the marginal difference between the first and tenth ranked model is comparably small.

http://epublications.bond.edu.au/ijbf/vol9/iss1/3

22

Trück and Liang: Forecasting volatility in the gold market

Overall, as could be expected, the out-of-sample errors are higher than the comparable numbers for the in-sample results. For MSE, there is an average increase by approximately 30 percent while for MAE, the average errors increase by approximately 80 percent in comparison to the in-sample results for the third subperiod. Again choosing the MAPE statistics provides slightly different findings and suggests the HM model as the best model followed by the OLS model. Once again the RW model performs worst. According to the U statistic, all models performed better than the RW model. Since U is based on the MSE measure, the results suggest the MA(40) followed by the EWMA model as being most appropriate. Generally the differences between the results for all models are rather small, since the values for Theil’s U range from 0.70 to 0.78. Results are confirmed by the DM test indicating that all models are significantly better than the benchmark RW model at the 0.01 level of significance. Considering the different model types we find that MA models performed clearly better in out-of-sample than in-sample forecasting, in particular the MA(40) model. Yet again there was no clear-cut outperformance of this model class with the MA(20) ranking fourth and the MA(120) ranking eighth. The HM model was the second worst according to the MSE, but it is one of the better models according to the MAE and MAPE. For the GARCH models it is noteworthy that the TARCH model, which provided superior in-sample fit for all three sub-periods, yielded one of the worst results for out-of-sample forecasting. On the other hand, in terms of the magnitude of MSE and MAE, the difference with the best model was still rather small. The ARMA models rank second to eleventh across the different measures indicating the importance of the right choice of the order of the coefficients. In summary, we conclude that there are only small differences with respect to the out-of-sample forecast performance between the considered models. The MA(40) could be considered the best model based on the MSE and U measures. Other models that have performed well are the ARMA(1,1) and the EWMA model. Furthermore, despite their generally good performance in the in-sample periods, for the considered out-of-sample period the GARCH models did not perform that well. In particular the TARCH model, that was the clear winner when in-sample volatility predictions were

Produced by The Berkeley Electronic Press, 2012

23

International Journal of Banking and Finance, Vol. 9, Iss. 1 [2012], Art. 3

considered, only ranked between 9 and 13 across the measures. Overall, there are no significant differences between the models and the rankings based on each performance measure are quite different. We conclude that, for the out-of-sample forecasting, it is hard to choose an overall winner. We will now extend our analysis by examining the different models with respect to risk quantification. In particular, we investigate and report their performance in forecasting Value-at-Risk (VaR). 4.3 Value at risk Analysis In this section, we examine the proposed models with respect to adequate VaR quantification in an out-of-sample forecasting study. For a given portfolio, probability and time horizon, VaR is defined as a threshold value of the probability that the markto-market loss of the portfolio over the given time horizon exceeds this value at a given probability level. In or analysis, following Giot and Laurent (2001, 2003), we evaluate the VaR forecasts from the perspective of both long and short position traders. The empirical models discussed in the previous sections are used to estimate determined as VaR = Zα ���� . In our analysis we consider one day 95% and 99% VaR � and forecast the volatility ���� . The VaR for a given level of significance α can then be �

and, thus, set α equal to 0.05 and 0.01.

The corresponding failure rates and VaR violations are then computed by comparing the one period ahead forecasts of VaR with the actual observed returns in the out-of-sample period with 127 observations. We define the number of violations for long traders as being equal to the number of times the negative observed return on a particular day exceeds the one-day-ahead VaR forecasts. Correspondingly, the number of violations for a short position is the number of times the positive return is larger than the determined VaR forecast for that particular day. The failure rate is the percentage of violations occurring in the out-of-sample period. If the model has been specified correctly, the failure rate should approximately be equal to the theoretical number of exceptions at the chosen VaR level, see e.g. Kupiec (1995); Christoffersen (1998); Christoffersen and Diebold (2000) or Hull (2007). The results for the calculated VaR forecasts for long and short positions in the gold market are provided in Table 7 and 8.

http://epublications.bond.edu.au/ijbf/vol9/iss1/3

24

Trück and Liang: Forecasting volatility in the gold market

Table 6: Out-of-sample Forecast Results for the Period July 1, 2008 to December 30, 2008

RW

HM

MA(20)

MA(40)

MA(120)

OLS

AR(5)

MAD(5)

ARMA(1,1)

EMWA

GARCH(1,1)

GARCH-M

TARCH

MSE Rank RMSE Rank MAE Rank MAPE Rank Theil U Rank DM Rank

172.870 13 13.148 13 8.124 13 10101.6 12 1.000 13 t-value p-value

105.245 12 10.259 12 5.433 2 2165.3 1 0.780 12 -1.764 12 0.040

84.649 4 9.200 4 5.966 12 10040.5 11 0.700 4 -2.432 5 0.008

83.291 1 9.126 1 5.722 9 9429.3 9 0.694 1 -2.457 3 0.008

87.972 8 9.379 8 5.401 1 7006.9 5 0.713 8 -2.281 9 0.012

101.224 11 10.061 11 5.654 5 3936.8 2 0.765 11 -1.997 11 0.024

86.081 5 9.278 5 5.673 6 6460.6 4 0.706 5 -2.399 6 0.009

90.080 10 9.491 10 5.546 3 5060.6 3 0.722 10 -2.257 10 0.013

84.249 3 9.179 3 5.681 7 8460.3 8 0.698 3 -2.437 4 0.008

83.812 2 9.155 2 5.844 10 9513.5 10 0.696 2 -2.459 2 0.008

86.942 7 9.324 7 5.569 4 7889.5 7 0.709 7 -2.308 8 0.011

86.352 6 9.293 6 5.685 8 7747.1 6 0.707 6 -2.327 7 0.010

88.238 9 9.394 9 5.855 11 13112.8 13 0.714 9 -2.726 1 0.003

Table 6 Out-of-sample forecast results for the period July 1, 2008 to December 30, 2008. Results are reported for examined models and the considered performance evaluation measures MSE, RMSE, MAE, MAPE, Theil U as well as results for Diebold-Mariano test with RW model as benchmark.

Produced by The Berkeley Electronic Press, 2012

25

International Journal of Banking and Finance, Vol. 9, Iss. 1 [2012], Art. 3

Given the confidence levels of 95% and 99% and a total of 127 days during the outof-sample period, one would expect approximately 6.35, respectively, 1.27 VaR exceptions. To test for the appropriateness of the considered VaR models, We apply a test that is based on the actual number of observed exceptions versus the expected number of exceptions, see e.g. Hull (2007). The test uses a binomial distribution such that given a true probability p of an exception, the probability of the VaR level being exceeded m or more days is:

k =m

∑ k !( n − k )! p

n

n!

k

(1 − p ) n − k
(22)

A similar reasoning applies to the case where the number of VaR violations m is lower than the expected number of exceptions. The probability of m or less exceptions is:

∑ k !( n − k )! p k =0

m

n!

k

(1 − p ) n − k
(23)

Based on these quantities it is easy to derive p-values for a correct VaR model specification given the number of exceptions that were actually observed. We find that the random walk model performs rather poorly both for the 95% and 99% VaR. For the long position, we observe 18, respectively 16 VaR exceptions corresponding to a failure rate of 14.2% and 12.6% that is substantially higher than the expected 5% and 1% under the assumption of a correct model specification. Similar results are obtained for holding a short position where the fraction of VaR exceptions is approximately 11% and 9.4%, respectively. Thus, as indicated by the pvalues, for both 95% and 99% VaR levels, the model is significantly rejected. While most of the models provide clearly less VaR violations than the RW model, only few of them are not rejected by the test for at least one of the two considered confidence levels. The HM and OLS model also significantly underestimate the risk, and yield too many exceptions for both long and short positions in particular at the 0.01 level. On the other hand, the three MA models yield a very small number of VaR violations, but the estimates are too conservative. As indicated in Table 7, for the long position, each MA model only yields one exception

http://epublications.bond.edu.au/ijbf/vol9/iss1/3

26

Trück and Liang: Forecasting volatility in the gold market

Tables 7: Out-of-Sample Results for Periods 2
RW HM AR(5) MAD(5) ARMA(1,1) EWMA GARCH(1,1) GARCH-M TARCH MA(20) MA(40) MA(120) OLS Long Position α = 5% (expected violations 6.35) Violations 18 12 1 1 1 9 3 5 2 2 2 2 4 Failure rate 0.142 0.094 0.008 0.008 0.008 0.071 0.024 0.039 0.016 0.016 0.016 0.016 0.031 p-value 0.0002 0.0113 0.0114 0.0114 0.0114 0.1046 0.1163 0.3860 0.0442 0.0442 0.0442 0.0442 0.2338 Long position α = 1% (expected violations 1.27) Violations 16 Failure rate 0.126 p-value 0.0000

8 0.063 0.000

0 0.000 0.2790

0 0.000 0.2790

1 7 2 0.008 0.055 0.016 0.6370 0.0000 0.1352

4 0.031 0.0093

1 0 0.008 0.000 0.6370 0.2790 EWMA

1 0.009 0.6370

1 2 0.008 0.016 0.6370 0.1352

Tables 8: Out-of-Sample Results for Period 3
RW HM OLS AR(5) MAD(5) MA(20) MA(40) MA(120) Short position α = 5% (expected violations 6.35) Failures 14 12 0 0 2 9 1 1 Failure rate 0.110 0.094 0.000 0.000 0.016 0.071 0.008 0.008 p-value 0.0018 0.0113 0.0015 0.0015 0.0044 0.1046 0.0114 0.0114 Short position α = 1% (expected violations 1.35) Failures 12 9 Failure rate 0.094 0.071 p-value 0.0000 0.0000 ARMA(1,1) GARCH(1,1) GARCH-M TARCH

0 0.000 0.0044

0 0.000 0.0044

1 0.008 0.0114

0 0.000 0.0044

1 0.008 0.0114

0 0.000 0.2790

0 0.000 0.2790

1 6 1 0.008 0.047 0.008 0.6370 0.0003 0.6370

0 0.000 0.2790

0 0.000 0.2790

0 0.000 0.2790

0 0.000 0.2790

0 0.000 0.2790

1 0.008 0.6370

Table7 & 8: Number of VaR violations for the considered models assuming a short position in gold. Results are reported for 1-day 95% and 99%-VaR. The p-values refer to a test based on Hull (2007) where the null hypothesis is an appropriate specification of the VaR model. Bold letters indicate models that were not rejected at the 5% significance level.

Produced by The Berkeley Electronic Press, 2012

27

International Journal of Banking and Finance, Vol. 9, Iss. 1 [2012], Art. 3

at the 95% VaR level leading to a rejection of the models even at the 0.10 significance level. Almost the same results are obtained for holding a short position in the gold market where the 95%-VaR estimates are also too conservative, so all MA models are rejected. Note however, that the models are not rejected for the 99%-VaR level since only a very small number of exceptions are expected at this level. Similar results are obtained for the ARMA, EWMA and two models with conditional variance GARCH(1,1) and GARCH-M model. These models only yield two exceptions at the 95% level and zero or one exception at the 99% level for a long position: for a short position, only the GARCH(1,1) model yields one exception at the 95% confidence level. The VaR estimates of these models are too conservative for the considered time period such that all models are rejected at the 5% significance level. The MAD(5) model gives too many exceptions at the 95% confidence level for a long position in gold, while it performs reasonably well at the 99% level for short positions. The best results – at least for long positions - are obtained for the AR(5) model and again for the threshold conditional volatility TARCH model. These models seem to provide adequate one-day-ahead risk forecasts for long positions and cannot be rejected for any of the considered confidence levels. Considering short positions, the models seem to provide estimates that are overly conservative and yield only one exception at the 95% and no exception at the 99% confidence level. Still, given the reasonable performance of the AR(5) and GARCH models for long positions, they could be considered as being most appropriate in terms of providing VaR forecasts. Overall, we conclude that there was no clear winner with respect to providing one-day ahead Value-at-Risk forecasts.

5. Summary and Conclusions In this paper we investigate the modelling of volatility dynamics of gold market returns in London. Gold markets are usually considered as a safe haven and investments into this class of assets have been very popular, in particular, since the global financial crisis. Therefore, appropriate models for volatility dynamics in these markets are of great interest to both investors and hedgers. While there are a number

59

http://epublications.bond.edu.au/ijbf/vol9/iss1/3

28

Trück and Liang: Forecasting volatility in the gold market

of recent studies examining volatility and Value-at-Risk (VaR) measures in financial and commodity markets, none of them focuses in particular on the gold market. Compared to the numerous studies on volatility modelling and forecasting focused on equity and commodity markets in general, we provide a pioneering study on the volatility of this important market. We contribute to the literature by using a large number of statistical approaches in order to model and forecast the daily volatility and Value-at-Risk in the gold spot market. Hereby, we distinguish between different time horizons including a sub-period of continuously but only slightly increasing gold prices, a sub-period of substantially increasing gold prices and, finally, a sub-period of high volatility in the gold market. Both in-sample and out-of-sample forecasts are evaluated using appropriate forecast evaluation measures. For in-sample forecasting, the class of TARCH models provided the best results among the tested models. Interestingly, the performance of a GARCH (1,1) model, that is generally supported by empirical studies for volatility modelling in financial markets (Akgiray, 1989; Franses and van Dijk, 1996), was only ranked in the middle of all models in our study. For out-of-sample forecasting, results were not that clearcut and the order and specification of the models was found to be an important factor in determining the model’s performance. VaR for traders with long and short positions were evaluated by comparing actual VaR exceptions to theoretical rates. For this task a simple AR as well as a TARCH model performed best for the out-ofsample period. We also find that most models were able to significantly outperform a benchmark random walk model both in the in-sample and the out-of-sample forecasting. However, none of the considered models performed significantly better than the rest with respect to all of the considered criteria. The out-of-sample period from July to December 2008 that has been tested in this study was one of the most volatile periods in the history of financial markets. As a result, the behaviour of the daily returns might be significantly different to previous periods and, also, possibly future periods. Thus, models that perform well in the considered out-of-sample period may well underperform in future periods, particularly when market conditions change. Second, though the study attempts to comprehensively investigate the volatility in the gold market by the means of using various models, it still only covered a small number of models available in this area.

Produced by The Berkeley Electronic Press, 2012

29

International Journal of Banking and Finance, Vol. 9, Iss. 1 [2012], Art. 3

For example, for models with conditional volatility, only three of the most widely used GARCH models were considered, leaving out a huge number of other GARCH model extensions. The flaws of VaR as a measure of risk along with the effectiveness of alternative risk measures such as expected shortfall, have been pointed out in the literature by e.g. Artzner et al. (1999). We leave the investigation of these issues to future work. Author information: Stefan Trück is a Professor of Finance in the Department of Applied Finance and Actuarial Studies and Co-Director of the Centre for Financial Risk at Macquarie University. Email: Stefan.trueck@mq.edu.au. Kevin Liang is a graduate student of Macquarie University. He has extensive professional experiences in the finance industry and is currently working as a credit risk analyst. Email: kzyliang@yahoo.com.au.

References Akgiray, V. (1989). Conditional heteroskedasticity in time series of stock returns: evidence and forecasts. The Journal of Business, 62(1): 55-80. Artzner, P., Delbaen, F., Eber, J.-M., & Heath, D. (1999). Coherent measures of risk. Mathematical Finance, 9(3): 203-228. Bollerslev, T. (1986). Generalized autoregressive conditional heteroskedasticity. Journal of Econometrics, 31(3) : 307-327. Bollerslev, T., Chou, R. Y., & Kroner, K. F. (1992). ARCH modeling in finance: A review of the theory and empirical evidence. Journal of Econometrics, 52(12) : 5-59. Brooks, C. (1998). Predicting stock index volatility: can market volume help? Journal of Forecasting, 17(1) : 59-80. Brooks, C., & Persand, G. (2002). Models choice and value at risk performance. Financial Analysts Journal, 58(5) : 87-97. Brooks, C., & Persand, G. (2003). Volatility forecasting for risk management. Journal of Forecasting, 22(1) : 1-22. Canarella, G., & Pollard, S. K. (2008). Modelling the volatility of the London gold market fixing as an asymmetric power ARCH. The Journal of Applied Finance, 14(5) : 17-43. Chou, R. Y. (1988). Volatility persistence and stock valuation: some empirical evidence using GARCH. Journal of Applied Econometrics, 3(4) : 279-294. Christoffersen, P. F. (1998). Evaluating interval forecasts. International Economic Review, 39(4) : 841-862.

http://epublications.bond.edu.au/ijbf/vol9/iss1/3

30

Trück and Liang: Forecasting volatility in the gold market

Christoffersen, P. F., & Diebold, F. X. (2000). How relevant is volatility forecasting for financial risk management. Review of Economics and Statistics, 82(1) : 1222. Danielsson, J., & de Vries, C. G. (2000). Value-at-risk and extreme returns. Annales d'Economie et de Statistique(60) : 11. Diebold, F. X. (1998). Elements of Forecasting. Cincinnati, Ohio: South Western College Publishing. Diebold, F. X., & Mariano, R. S. (1995). Comparing predictive accuracy. Journal of Business and Economic Statistics, 13(3) : 253-263. Dimson, E., & Marsh, P. (1990). Volatility forecasting without data-snooping. Journal of Banking and Finance, 14(2-3) : 399-421. Ding, Z., Granger, C. W. J., & Engle, R. F. (1993). A long memory property of stock market returns and a new model. Journal of Empirical Finance, 1(1) : 83-106. Engle, R. F. (1982). Autoregressive conditional heteroskedasticity with estimates of the variance of United Kingdom inflation. Econometrica, 50(4) : 987-1007. Engle, R. F., Lilien, D. M., & Robins, R. P. (1987). Estimating time varying risk premia in the term structure: the ARCH-M model. Eonometrica, 55(2) : 391407. Franses, P. H., & van Dijk, D. (1996). Forecasting stock market volatility using (nonlinear) GARCH models. Journal of Forecasting, 15: 229-235. French, K. R., Schwert, G. W., & Stambaugh, R. F. (1987). Expected stock returns and volatility. Journal of Financial Economics, 19(1) : 3-29. Giot, P., & Laurent, S. (2003). Market risk in commodity markets: a VaR approach. Energy Economics, 25(5) : 435-457. Giot, P., & Laurent, S. (2003). Value-at-risk for long and short trading positions. Journal of Applied Econometrics, 18(6) : 641-663. Glosten, L. R., Jagannathan, R., & Runkle, D. E. (1993). On the relation between the expected value and the volatility of the normal excess return on stocks. The Journal of Finance, 48(5) : 1779-1801. Hansen, P. R., & Lunde, A. (2005). A forecast comparison of volatility models: does anything beat a GARCH (1,1). Journal of applied econometrics, 20(7) : 873889. Henriksson, R. D., & Merton, R. C. (1981). On market timing and investment performance. II. Statistical procedures for evaluating forecasting skills. Journal of Business, 54(4) : 513-533. Hentschel, L. (1995). All in the family, nesting symmetric and asymmetric GARCH models. Journal of Financial Economics, 39(1) : 71-104. Hull, J. C. (2007). Risk management and financial institutions: Pearson Prentice Hall. Jarque, C. M., & Bera, A. K. (1980). Efficient tests for normality, homoskedasticity and serial independence of regression residuals. Economics Letters, 6(3) : 255259. Jorion, P. (1996). Risk2: Measuring the risk in value at risk. Financial Analysts Journal, 52(6) : 47-56. Jun, J. H. (2009) Global financial crisis and gold market. Working Paper. Kupiec, P. (1995). Techniques for veryfying the accuracy of risk management models. Journal of Derivatives, 3(2) : 73-84. McMillan, D., Speight, A., & Apgwilym, O. (2000). Forecasting UK stock market volatility. Applied Financial Economics, 10(4) : 435-448.

Produced by The Berkeley Electronic Press, 2012

31

International Journal of Banking and Finance, Vol. 9, Iss. 1 [2012], Art. 3

Morales, L. (2008). Volatility spillovers on precious metals markets: the effects of the Asian crisis. Paper presented at the European Applied Business Research Conference (EABR), Salzburg, Austria. J.P. Morgan., RiskMetrics Monitor, Fourth Quarter, 1995. Nelson, D. B. (1991). Conditional heteroskedasticity in asset returns: a new approach. Econometrica, 59(2) : 347-370. Sadorsky, P. (2006). Modeling and forecasting petroleum futures volatility. Energy Economics, 28(4) : 467-488. Stock, J. H., & Watson, M. W. (1998). A comparison of linear and non-linear univariate models for forecasting macroeconomic time series. NBER Working Paper, 6607. Tully, E., & Lucey, B. M. (2007). A power GARCH examination of the gold market. Research in International Business and Finance, 21(2) : 316-325.

http://epublications.bond.edu.au/ijbf/vol9/iss1/3

32

Similar Documents

Free Essay

Student

...Revision of Critical essay *Introduction In today's society there is a lot of pressure on students academically to have a good performance and with that comes a lot of stress. Some students find a way to try to balance their hectic school life style whether it be some kind of recreational activity. One of those activities is sports and whether it can make a better student. I believe that yes it can increase your performance academically because it teaches you skills such as focus, fitness and communication with others. In the article “do athletes make better students, Natalie Gil written for the guardian.com. Natlie Gil claims that studies show that doing both can benefit studies and sports performance, providing motivation and preparation. Natalie Gil also goes on to state that it helps organization and pervents procrastination and that being fit alters students mood in a good way claiming a healthy body is a healthy mind. Lastly, Natalie Gil goes on to show evidence that it also helps with communication and team work whether at school or later in landing a career. Pathos Natalie Gil Appeals to the stress and desire to succeed in today's world as students upcoming in today's society. She also uses the points or appeal to support her view or stance on the subject that athletes do make better students and that this will lead to success not only in their academic life but also in their career choice Logos ......

Words: 616 - Pages: 3

Premium Essay

Student

...are important to be included in the evaluation of teaching effectiveness. These factors are as the criteria for the evaluating of educational effectiveness. Some of these factors still work as a criterion for the evaluation process. While, the other factors have to be excluded from the evaluation and not to be given as much weight. Therefore, the main goal of this study is to ask administrators about which items still valid until the now and have to be included in the evaluation process and which of these items are invalid to be an evaluation criterion. This article also offers the main sources of data for evaluation of faculty performance as one of the important components of evaluation of educational effectiveness. There sources are students’ evaluation tools, teaching portfolios, classroom visitation reports, and scholarship activities. These sources offer significant information about the faculty performance and consequently they will contribute significantly in assessing and evaluating the teaching effectiveness. There are some items of evaluation have to be included and be given more weight in any evaluation process of the educational effectiveness because they have a significant relation to the success of the evaluation process. These items are currency in field, peers evaluation, classroom visits, professors preparations. While, there are some items have to be excluded because they do not contribute in success of evaluation of teaching effectiveness...

Words: 325 - Pages: 2

Free Essay

Student

...SOX testing, I was also assigned to assist building the Compliance Universe for the whole organization. I appropriately allocated my time and energy to these two projects, so that I completed most of my work in a high quality and on a timely basis. I am a dedicated team player who loves communicating with people. I interviewed Hologic’s employees to understand key business processes, joined all the staff meetings and presented my ideas and achievements to the team, collaborated with colleagues to work on other projects to meet the deadline. I am also a person with great research and analytical skills. I used CCH, FASB Codification and some other information sources to finish my cases in academic study. Even though I am an international student, I believe that I am better for this position than anyone else. Companies like Signiant need global perspective people. I majored in International economy and trade during undergraduate study. I have knowledge about foreign currency, international transactions and taxes. All I need is a chance to learn and contribute in a fast-paced company like Signiant. The enclosed resume briefly summarizes my educational background and experiences, I would like to meet with you for an interview during which I can fully express my capacity and desire to work for Signiant. In the meantime, if you need any additional information, please contact me by phone at 781-502-8582 or via e- mal at liulezi2012@hotmail.com Thank you for your time and...

Words: 319 - Pages: 2

Free Essay

Student

...THE RATE OF INVOLVEMENT OF KPTM KL’S STUDENTS IN SPORTS AT THE COLLEGE Prepared by : MUHAMMAD AEZHAD BIN AZHAR CVB130724387 MUHAMMAD FARHAN BIN ABDUL RAHMAN CVB130724287 RAHMAN MUSTAQIM BIN KHOSAIM CVB130724279 MUHAMMAD AIMAN BIN MOHD HUSNI CVB130724388 Prepared for : Madam Jaaz Suhaiza Jaafar Submitted in partial fulfillments of the requirement of the 106km course. TABLE OF CONTENTS NUMBER | CONTENTS | PAGES | 1. | ACKNOWLEDGEMENT | 3 | 2. | INTRODUCTION | 4 | 3. | OBJECTIVES | 5 | 4. | METHODOLOGY | 6-7 | 5. | GRAPH | 8-11 | 6. | CONCLUSION | 12 | 7. | APPENDIX TABLE | 13 | 8. | APPENDIX | 14-17 | ACKNOWLEDGEMENT First of all,we really want to thankful to Madam Jaaz Suhaiza Jaafar because allowed me to do this mini project until we’ve successfully completed it.We want thankful too because madam helped us a lot such as give instructions or order how to make it properly done until we’ve finished it. If we didn’t get help from madam,its really hard to us for completed it in a short time. We also want to very thankful too all our 50 respondents which all of them its from KPTM KL students who was in diploma,degree or professional. They all was nice and very friendly with us and nobody refuse to give a little time to fill up our questionnaire. We really want to wish thanked you so much because without them we can’t finished our mini project. Last but not least,thank you so much too our...

Words: 2116 - Pages: 9

Free Essay

Student

...Study of Asia-Pacific MBA Programs Bloomberg Business week posted an article on March 17th 2014 titled, Elite Business Schools Hike Tuition for the Class of 2016. This article draws a comparison between tuition costs for the class of 2015 for selected US MBA programs and the class of 2016. Tuition costs are increasing more and more every year, for this reason looking at other alternatives may be more cost effective. The following study provides and interpretation of tuition cots both local and foreign in the Asia-Pacific region. From this study we can see the comparison between tuition costs and starting salaries. We can also see other deciding factors such as admission requirements. Finally this study provides a recommendation for an MBA program in the Asia-Pacific region. Please note Table 1.1 listing the study’s programs with their correlating graph ID. Table 1.1 Business School | Graph ID | Lahore University of Management Sciences | LUMS | Indian Institute of Management (Calcutta) | IIMC | University of New South Wales (Sydney) | UNSW | Indian Institute of Management (Bangalore) | IIMB | Curtin Institute of Technology (Perth) | CIT | Massey University (Palmerston North, New Zealand) | MU | University of Queensland (Brisbane) | UQ | University of Adelaide | UA | Monash Mt. Eliza Business School (Melbourne) | MMEBS | Melbourne Business School | MBS | Royal Melbourne Institute of Technology | RMIT | Macquarie Graduate School of......

Words: 3907 - Pages: 16

Premium Essay

Student

...playing a basic rule in the education, and the government was searching for a solution to eliminate this phenomenon. They found that establish public schools overall the states will improve a lot of the poor income people to be introduced in the educational field, and over the years will produce community with cultured educated society. The education is varies in all levels, starting from preschool reaching to postgraduate like masters and doctoral degree. The insurance of improvement in education that any non U.S graduate must have multiple exams prior to admission e.g. TOEFL, ILETS, GRE, GMAT. Nowadays there are gradual increase in the numbers of international students want to continue their educations in United States. The improvement of the education in United States is very obvious and attracts the students worldwide, and they release a lot of plans in progress. All the opportunities social, health, economic, academic will depend on the basic structure...

Words: 306 - Pages: 2

Free Essay

Student

...Retention(n), retain verb (used with object) the ​continued use, ​existence, or ​possession of something or someone:Two ​influential ​senators have ​argued for the retention of the ​unpopular ​tax.The retention of ​old ​technology has ​slowed the company's ​growth.​water/​heat retention Particularly(adv) Especially(adv) Deter(v) to make someone less likely to do something, or to make something less likely to happen caydırmak, vazgeçirmek, yıldırmak Perception(n) BELIEF [C]› what you think or believe about someone or something algılama, sezgi, görme The public perception of him as a hero is surprising. NOTICE [U] the ability to notice something fark etme, farkına varma, tanıma, görme Alcohol reduces your perception of pain. Conationimpulse Unanimous agreed by everyoneoy birliği ile üzerinde uzlaşılan; herkesçe kabul edilen; genel kabul görenThe jury was unanimous in finding him guilty. unanimity     /ˌjuːnəˈnɪməti/ noun [U]› when everyone agrees about somethinggenel/toplumsal uzlaşı; oy birliği ile anlaşma; genel kabul; fikir birliğiunanimously adverb›oy birliği ile kabul edilmişThe members unanimously agreed to the proposal. dissonancenoun [U]  UK   /ˈdɪs.ən.əns/  US   /ˈdɪs.ə.nəns/      › specialized music a ​combination of ​sounds or ​musical ​notes that are not ​pleasant when ​heard together:the ​jarring dissonance of Klein's ​musical ​score› formal ​disagreement dissonant adjective UK   /ˈdɪs.ən.ənt/  US   /ˈdɪs.ə.nənt/ specializedor formal ›a dissonant......

Words: 335 - Pages: 2

Premium Essay

Student

...Student Handbook 2015/2016 www.praguecollege.cz Table of Contents Introduction Message from the Director Mission, Vision and Values Why study at Prague College Admissions A short guide to Prague College qualifications English for Higher Education Foundation Diploma in Business Foundation Diploma in Computing Foundation Diploma in Art & Design Professional Diplomas in Business Professional Diplomas in Computing Higher National Diploma BA (Hons) International Business Management BA (Hons) International Business Management (Flexible Study Programme) BA (Hons) Business Finance & Accounting BA (Hons) Graphic Design BA (Hons) Fine Art Exp. Media BSc (Hons) Computing BA (Hons) Communications & Media Studies MSc International Management MSc Computing Accreditation & Validation UK/Pearson Credit system Transfer of credits Student support Accommodation Study Advising and Support Financial support Visas for foreign students Scholarships Benefits for students Study abroad Internships Assistance in employment Counselling Centre Student Resources Computer labs Online Learning Centre (Moodle) Prague College email Physical library Digital Library ISIFA Images Textbooks and class materials Graphic Design/Interactive Media/Fine Art materials and costs Personal computers Message boards and digital signs Newsletters Open lectures, seminars and events Student ID cards Centre for Research and Interdisciplinary Studies......

Words: 27092 - Pages: 109

Free Essay

International Student

...[pic] TOPIC: INTERNATIONAL STUDENTS’ ATTITUDES ABOUT HIGHER EDUCATION IN THE UK Student: Pham Trang Huyen My Student ID: 77142444 10 weeks Pre-sessional course December, 2013 List of content Abstract 3 1. Introduction 4 2. Literature review 5 2.1. Higher Education in the UK 5 2.2. Teacher-student relationships and the quality of teaching 5 2.3. Different learning styles 6 2.4. Group work 7 2.5. Financial issues 8 3. Methodology 9 4. Results 10 5. Discussion 14 6. Conclusion 16 List of References 17 Appendix 19 Abstract Higher education is a competitive business which produces huge benefits for the UK economy. This paper reveals international students’ attitudes about UK higher education and focuses on direct factors which can affect students’ opinions. Reports of international students’ attitudes already carried out in Leeds Metropolitan University are analyzed and the main findings are emphasized. A total of eighteen international students interviewed provided data on their experience in UK education that involves the challenges they have faced and what they have achieved. The project concludes that not only UK tuition fees but also the quality of education can affect international students’ decision to study in the UK. Therefore measures should be taken......

Words: 3732 - Pages: 15

Free Essay

Working Student

...INTRODUCTION Many students of HRM in Taguig City University work part-time Employment during school could improve grades if working promotes aspects that correspond with academic success, such as industriousness or time management skills, or instead reduce grades by reducing time and energy available for school work. Otherwise, working might be associated with academic performance, yet not directly influence it, if unobserved student differences influence both labor supply and grades. Unmotivated students might neither work for pay nor receive good grades because they put little effort into the labor market or school. In contrast, HRM students uninterested in academics might work long hours that would otherwise have been devoted to leisure. Students might misjudge the link between college achievement and future earnings when making labor supply decisions. If so, obtaining a consistent estimate of how such decisions affect academic performance is prospectively important for policy consideration. Some of HRM students in Taguig City University Students are more likely to work than they are to live on campus, to study full time, to attend a four-year college or university, or to apply for or receive financial aid. Students work regardless of the type of institution they attend, their age or family responsibilities, or even their family income or educational and living expenses. Most HRM students at Taguig City University face many challenges in their already busy everyday lives...

Words: 2898 - Pages: 12

Free Essay

Student Adversity

... Adversity allows an individual to develop a sense of discipline, as well as encouraging individuals to exercise their mind to confront a problem or conflict. Specifically, students who encounter hardships are more inclined to try harder, which promotes competition within the school. Although adversity may be beneficial towards some students, challenges can be detrimental for students who lack confidence. For instance, some students develop a mentality of despair; they believe that if one has to work hard, then the person does not have the natural ability for the assignment. Based on the effects of adversity aforementioned, I believe that students can both benefit from the obstacles faced in school with the proper mentality or the effects could be hindering. Students face adversity every day, regardless of how transparent the obstacle may be; some problems may not be as evident as others. According to Carol S. Dweck, author of Brainology, all students face adversities throughout their high-school career, specifically, the challenge of overcoming a fixed mindset. In this excerpt, “The belief that intelligence is fixed dampened students’ motivation to learn, made them afraid of effort, and made them want to quit after a setback”, Carol portrays the illusion that students have over intuitive intelligence (Dweck 2). Students who share this belief of a...

Words: 1029 - Pages: 5

Free Essay

Student Handbook

...Student Handbook (Procedure & Guideline) for Undergraduate Programmes 2014 Revised: April 2014 UCSI Education Sdn. Bhd. (185479-U) VISION AND MISSION STATEMENT OF UCSI UNIVERSITY VISION STATEMENT To be an intellectually resilient praxis university renowned for its leadership in academic pursuits and engagement with the industry and community MISSION STATEMENT  To promote transformative education that empowers students from all walks of life to be successful individuals with integrity, professionalism and a desire to contribute to society  To optimize relationships between industry and academia through the provision of quality education and unparalleled workplace exposure via Praxis Centres  To spearhead innovation in teaching and learning excellence through unique delivery systems  To foster a sustainable culture of research, value innovation and practice, in partnership with industries and society  To operate ethically at the highest standards of efficiency, while instilling values of inclusiveness, to sustain the vision for future generations 2 UCSI Education Sdn. Bhd. (185479-U) Graduate Attributes Getting a university degree is every student‟s ultimate dream because it opens doors to career opportunities anywhere in the world. A university degree is proof of one‟s intellectual capacity to absorb, utilize and apply knowledge at the workplace. However, in this current competitive world, one‟s knowledge and qualifications...

Words: 28493 - Pages: 114

Premium Essay

Student Policy

...Student Academic Policies Computer Usage: Sullivan University Systems (SUS) provides computer networking for all staff, students and anyone else affiliated with the university community. Sullivan University will provide a platform that is conducive for learning while maintain and respecting the user privacy. Users are authorized to use the accounts only. Passwords should be protected, please keep the confidential (Computer Usage. (2012) Sullivan University. Student Handbook 2012-2013, pp. 12-14.). While using the SUS users have a responsibility and are expected to follow some key rules: 1. Do not abuse the equipment 2. Computers must be used for course work 3. No unauthorized down loading 4. At no time will user install software of any kind Disciplinary action for violations of the Computer usage of policy will be enforced and are as follows: 1. Loss of computer privileges 2. Disconnection from the network 3. Expulsion 4. Prosecution The Compute usage policy is standard and pretty straight forward. The statement lets students know what is and is not proper usage. What I would have like to have seen is a social media portion in the usage policy. Academic Integrity: Cheating and Plagiarism is a violation of the University’s Academic Integrity Policy. All students are expected to submit their own work. Penalties for those who are found guilty of cheating may include: (Academic Integrity. (2014, January 1) Sullivan University. Sullivan University 2014......

Words: 320 - Pages: 2

Premium Essay

Student Satisfaction

...between the quality of school facilities and student...

Words: 2174 - Pages: 9

Premium Essay

Working Students

...performance of hiring working students Introduction While most students have parents that can support them, there are those students that need get what you call a “part-time job” to help their parents that can’t support them all the way. However, being employed and being a student can be too much to a person. The business process outsourcing industry in the Philippines has grown 46% annually since 2006. In its 2013 top 100 ranking of global outsourcing destinations. Significance of the Study There are situations in the life when one must do what they can to achieve their dreams or help their families. Especially if dealt with financial difficulties and there is a need work while studying. They also need to deal with their everyday busy schedules. This research aims to help understand and discuss the issues and concerns of the employed students to benefit the following: Working Students – Being an employee and student at the same time takes a lot of hard work. It can be rigorous but also rewarding especially if you helped your parents. It can also be a good working experience for them for their future. This study will assist them to see the behaviors that help them achieve their professional skills. Scope and Limitations This is study is conducted at the LPU-Manila and the information is viewed only in the light of the particular student and his or her experience as working student. It does not reflect the view of the general working student population or that......

Words: 606 - Pages: 3