# a critical comparative analysis of adoption in Islam and the British value Research Paper Assignment Essay

Does Speculation of Futures of Food Commodities impact the Food Prices? Methodology Overview of the design Excessive speculation can sometimes be singled out as a possible factor that can be responsible for the fluctuations and volatility of the prices in commodities at different times of the year[1]. It positively increases the energy prices that in turn results to the increased food and commodity prices since for the production of the commodity, energy must be a key factor to be involve[2]. Volatility can generally be defined as the variances that are involved in assets returns, it cannot be directly observed. Since speculations are responsible for the fluctuations and volatility of the returns, we can therefore use a methodology that helps in analyzing this volatility[3]. The statistical model that is reputably known for modeling the volatility of returns of a product is the conditional heteroscedastic models. Sometimes the conditional heteroscedastic model is simple, but it normally involves the use of many parameters to sufficiently model the volatility of a commodity return. An extension of this conditional heteroscedastic model was proposed by Bollerslerv in 1986, the extension which is known as Generalized conditional heteroscedastic model (GARCH)[4]. Garch model is appropriate for the research over other models because of the few parameters it involves and the clustering of volatility where high shock are followed by high shocks followed by low shocks i.e. it responds slowly to the large isolated shocks of the return of a particular commodity[5]. The GARCH process it suitable for modeling the conditional variance that helps in capturing the time varying characteristics of price volatility. GARCH is preferred over OLS because the latter requires the error term of in order to be homoscedastic[6]. The idea behind the Garch model is that, the shocks of the return of a comodity are serially uncorrelated and they are always dependent[7]. The dependence of the shocks of the commodity return can be owed to a simple function of its past lagged values over time. For the research that is to be conducted to determine whether Speculation of Futures of Food Commodities impact the Food Prices, we will examine the stationarity of the series of the return of the commodity. This is to be done for three commodities (future and spot) in order to explore the implication of co-movement of the data. This will help in reaching a confirming the feasibility of the results from the determined correlation. To test for this stationarity in the return of a commodity in R programming, we will use the Dickey fuller test[8]. Consider the return of a commodity lagged once to be given by the equation If the coefficient of which is equals to 1 then the above process reduces to a process called random walk. Random walk processes are always known to be stationary[9]. For us to test for stationarity in the return of the commodity, we will use the Dickey Fuller test that has its hypothesis given by The null hypothesis is defined by This implies that there is no stationarity in the time series of the return of the commodity The alternative hypothesis is given by This implies that there is stationarity in the time series of the return of the commodity. To do this test for stationary in R programming we will use the function below to test for stationarity Where data represent the returns of the commodity in the time series[10] On rejecting the null hypothesis of non stationarity, we conclude that the time series of the return of the commodity is stationary; this implies that the Speculation of Futures of Food Commodities doesn t impact the Food Prices[11]. But if we fail to reject the null hypothesis of non stationarity, we conclude that the time series data for the return of the commodity is not stationary. This non stationarity can be as a result of Speculation of Futures of Food Commodities, hence we can confirm that the Speculation of Futures of Food Commodities impact the Food Prices[12]. If there is a correlation between spot price and future prices then it would be concluded that the speculation of futures will result to volatility in spot prices. Testing for arch effect After performing the dickey fuller test for stationarity, the efficiency of GARCH model would be tested by testing for arch effects. The following is the R-code and the R-output for testing the arch effects R-code ArchTest(logreturn,lags=1) In the process, an hypothesis for testing the arch effects will be formulated as shown below H0: there are arch effects H1: there are no arch effects From this, the resulting p-value would help us to reject or fail to reject the null hypothesis. If this p-value is less than 0.05, it will mean that the null hypothesis is to be rejected and conclude that there is a presence of the arch effects. In this case, this will have helped in giving assurance for the applicability of GARCH model in the research. MODEL BUILDING PROCESS In building the model of the volatility if the return for the commodity, the following 3 steps will be considered The model order identification will be carried out using the plot of the partial autocorrelation functions of the shocks[13]. Parameter estimation of the chosen model using the maximum likelihood method and the OLS methods to determine the the parameters[14] The last method is model checking and forecasting The best model is the one that will minimize the Akaike Information Criteria, i.e. the model that has the smallest value[15] Structure of the model Garch model is defined as follows Fitting GARCH model in R #importing the file called WHEATSPOT from Microsoft excel wheat.spot<-read.csv( wheatspot.csv , header=T) wheat.spot attach(wheat.spot) head(wheat.spot) #converting the data to time series ts(wheat.spot, start = c(1999,1), frequency = 12) #plotting the time series ts.plot(log.return) #garch command Garchfit.for.wheat.spot<-garchFit(.~garch(1,0), data=logreturns, trace=F,cond.dist= std ) #calling the garch command Garchfit.for.wheat.spot Output that we will produce in R when we run the above code for modeling GARCH Coefficient(s): mu omega alpha1 shape -7.0799e-05 5.2631e-04 2.2196e-01 3.8038e+00 Therefore the garch equation for the spot prices is given by Analyzing the relationship In analyzing the relationship between the data, spot price volatility and effect of future prices, two important aspects are to be considered. First is to determine whether the future trading price existence affects volatility? Second, is to analyze how the future trading affects volatility, if it really does affect volatility? That is, determine the relationship existing between information and volatility as the result of beginning of future trading[16]. To solve the first question, the conditional variance equation is augmented with a dummy variable by utilizing the value zero pre-futures and one post-futures. Hence the question becomes: DF in the equation represents a dummy variable. If the dummy is found to be statistically significant then it can be concluded that the future price has effect on the volatility of spot prices[17]. For the second question, the period taken for the study is divided into two periods relating to period for pre and post-future trading. This would allow for comparison of volatility exhibited before and after the beginning of future trading. How we know that only volatility is from speculation and not from market supply and demand In order to squarely isolate the impact of speculation, we will have to assume market elasticity of supply and demand[18]. In this way, the above model will be effective in determining the influence of speculation on food prices. Without the belief of price elasticities, it would be difficult to isolate futures as a significant driver of food prices. Therefore, the following test will be carried out First we will have to perform calculation for changes in inventory for speculation to be considered a driver of observed increase in price. Secondly, perform calculation on the elasticity of demand and supply required for the witnessed inventory changes to lead to increase in prices. Third, perform calculation for the no-speculation change associated with convenience yield. This is then compared to the actual[19]. From the above, we will be able to determine if speculation alone have impact on the commodity prices. How the time series for the returns of the commodity will be done in R ts(wheat.spot, start = c(1999,1), frequency = 12) ts.plot(log.return) Figure 1: Diagram sowing time series plot for the log return of commodities. We use the return and not prices since returns are easier to work and compute and they have many statistical properties compared to price[20]. This makes it advisable to use returns and not prices to compute the time series[21]. The easiest type of return to use is the continuously compounded return also called the log return which is denoted by Since volatility is not directly observable; the monthly volatility cannot be observed in the time series of the data since it is usually given as a single value[22]. Therefore to observe the volatility in a given month, it is in order to record the intra month data in terms of days and even weeks so that the volatility in that month can be observed[23]. Once the intra month data is recorded daily or weekly, it will be easier to compute volatility of the commodity return since the daily data in the month can give the variation that occurred in the month[24]. Therefore, it will be advisable to use the daily or weekly data rather than that of the month. This is because the data for the month is only a single value it will not be able to give the volatility in that given month[25]. Garch will then be used to model Volatility of these intra month volatility since volatility is the variance of the returns of the commodity in the given time series[26]. When the tests for the stationarity are performed using the dickey fuller test to check for the stationarity of the time series, the output statistics would be as shown below. R code for stationarity check #Importing data from excel data<-read.csv( data.csv , header=T) data attach(data) #Stationarity check code adf.test(data) Output in R Augmented Dickey-Fuller Test data: dataDickey-Fuller = XXXX, Lag order = X, p-value = 0.XXalternative hypothesis: stationaryInterpretation 0.05 is our critical value that we make comparison with i.e. it is 5% level of significance, we can also use 1% level of significance i.e. 0.01. If the p-value i.e. 0.XX will be less than 0.05 We reject the null hypothesis of non stationarity and conclude that the process is stationary. Therefore the Speculation of Futures of Food Commodities doesn t impact the Food Prices but if the p-value of 0.XX is greater than 0.05 we fail to reject the null hypothesis of non stationarity and conclude that the process is non stationary and therefore Speculation of Futures of Food Commodities impact the Food Prices. R codes for plotting graphs The function Plot (x, y, xlab= x , ylab= y ) gives scatter plot of y against x. Hist(x) produces histogram of x ts.plot(x) produces time series plot for data x acf(x) produces autocorrelation function for x pacf(x) produces partial autocorrelation function for x barplot(x) produces a bar graph of x pie(x) produce a pie chart of x OLS can be required if one would like to know the relationship between a dependent factor and other several independent factors in a given data[27]. R code for OLS regression Let y be the dependent variable and variables x1, x2 and x3 be the independent variables To do the regression of the dependent variables against the independent variables we will use the above code in R y<-c(enter the values of y here separated by coma) x<-c(enter the values of x here separated by coma) model<-lm(y~x1+x2+x3) summary (model) example y<-c(1,2,3,4,5,6,7,8,9,10,11) x1<-c(2,5,6,9,8,7,15,14,16,21,25) x2<-c(12,13,14,15,16,17,18,19,25,41,23) x3<-c(11,21,23,24,25,26,27,28,29,31,32) model<-lm(y~x1+x2+x3) summary(model) output in R Call : lm(formula = y ~ x1 + x2 + x3)(Intercept) -2.83926 1.87228 -1.516 0.1732 x1 0.27351 0.09328 2.932 0.0220 * x2 0.03824 0.05691 0.672 0.5232 x3 0.19523 0.10062 1.940 0.0935 . Residual standard error: 0.9549 on 7 degrees of freedom Multiple R-squared: 0.942, Adjusted R-squared: 0.9171 F-statistic: 37.88 on 3 and 7 DF, p-value: 0.0001071 Interpretation Y=-2.83926 + 0.27351??1 + 0.03824??2 + 0.19523 x3 Bibliography Bailer, Heiko Manfred. 2005. Robust Estimation of Factor Models in Finance. Order No. 3183337, University of Washington. http://search.proquest.com/docview/305383356?accountid=45049. Bunnag, Tanattrin. 2014. Volatility Analysis of International Tourist Arrival Growth Rates to Thailand using Garch and GJR Model. Journal of Environmental Management & Tourism 5 (1): 71-84. doi:http://dx.doi.org/10.14505/jemt.v5.1(9).06. http://search.proquest.com/docview/1628555215?accountid=45049. Conrad, Christian and Menelaos Karanasos. 2010. Negative Volatility Spillovers In The Unrestricted Eccc-Garch Model. Econometric Theory 26 (3): 838- 862. doi:http://dx.doi.org/10.1017/S0266466609990120. http://search.proquest.com/docview/212365852?accountid=45049. Daouk, Hazem and Jie Qun Guo. 2004. Switching Asymmetric Garch and Options on A Volatility Index. The Journal of Futures Markets 24 (3): 251-282 . http://search.proquest.com/docview/211217006?accountid=45049. Fong, W. M. and K. H. See. 2001. Modelling the Conditional Volatility of Commodity Index Futures as a Regime Switching Process. Journal of Applied Econometrics 16 (2): 133- 163. http://search.proquest.com/docview/218730553?accountid=45049. F?ªss, Roland, Dieter G. Kaiser, and Zeno Adams. 2007. Value at Risk, GARCH Modelling and the Forecasting of Hedge Fund Return Volatility. Journal of Derivatives & Hedge Funds 13 (1): 2-25. doi:http://dx.doi.org/10.1057/palgrave.jdhf.1850048. http://search.proquest.com/docview/232036370?accountid=45049. Grzywacz, Piotr and Krzysztof WoIyniec. 2011. Multi-Scale Volatility In Commodity Markets. Energy Risk, 08, 44-49. http://search.proquest.com/docview/883131487?accountid=45049. Kolade, Sunday Adesina. 2013. Modelling Stock Market Return Volatility: GARCH Evidence from Nigerian Stock Exchange. International Journal of Financial Management 3 (3): 37-46. http://search.proquest.com/docview/1478013012?accountid=45049. Kolade, Sunday Adesina. 2013. Modelling Stock Market Return Volatility: GARCH Evidence from Nigerian Stock Exchange. International Journal of Financial Management 3 (3): 37-46. http://search.proquest.com/docview/1478013012?accountid=45049. Levenbach, Hans. 1982. Time Series Forecasting using Robust Regression. Journal of Forecasting (Pre-1986) 1 (3): 241. http://search.proquest.com/docview/224777190?accountid=45049. Liu, Wei and Bruce Morley. 2009. Volatility Forecasting in the Hang Seng Index using the GARCH Approach. Asia ? Pacific Financial Markets 16 (1): 51-63. doi:http://dx.doi.org/10.1007/s10690-009-9086-4. http://search.proquest.com/docview/215208516?accountid=45049. Makhwiting, M. R., M. Lesaoana, and C. Sigauke. 2012. Modelling Volatility and Financial Market Risk of Shares on the Johannesburg Stock Xchange. African Journal of Business Management 6 (27): 8065-8070. doi:http://dx.doi.org/10.5897/AJBM11.2525. http://search.proquest.com/docview/1030941884?accountid=45049. McMillan, David G. and Alan E. H. Speight. 2004. Daily Volatility Forecasts: Reassessing the Performance of GARCH Models. Journal of Forecasting 23 (6): 449. http://search.proquest.com/docview/219199811?accountid=45049. Rodrigues, Paulo M. M. and Antonio Rubia. 2004. On the Small Sample Properties of Dickey Fuller and Maximum Likelihood Unit Root Tests on Discrete-Sampled Short-Term Interest Rates. St. Louis: Federal Reserve Bank of St Louis. http://search.proquest.com/docview/1698624756?accountid=45049. Rogers, Nina, Kimberly Winson-Geideman, and Imre Karafiath. THE IMPACT OF TRADING VOLUME ON REIT VOLATILITY USING THE GARCH MODEL. Journal of Real Estate Portfolio Management 20, no. 3 (2014): 167-177. http://search.proquest.com/docview/1672098827?accountid=45049. Shaikh, Imlak and Puja Padhi. 2014. Stylized Patterns of Implied Volatility in India: A Case Study of NSE Nifty Options. Journal of Indian Business Research 6 (3): 231. http://search.proquest.com/docview/1650590445?accountid=45049. Storti, Giuseppe. 2008. Modelling Asymmetric Volatility Dynamics by Multivariate BL- GARCH Models. Statistical Methods & Applications 17 (2): 251-274. doi:http://dx.doi.org/10.1007/s10260-007-0066-4. http://search.proquest.com/docview/213115833?accountid=45049. Wai-Ching Poon, Chee-Keong Choong, and Shah Habibullah Muzafar. 2005. Exchange Rate Volatility and Exports for Selected East Asian Countries: Evidence from Error Correction Model. ASEAN Economic Bulletin 22 (2): 144-159. http://search.proquest.com/docview/219629289?accountid=45049. Wen-Hsiu Kuo, Hsinan Hsu, and Chwan-Yi Chiang. 2005. Price Volatility, Trading Activity and Market Depth: Evidence from Taiwan and Singapore Taiwan Stock Index Futures Markets. Asia Pacific Management Review 10 (2). http://search.proquest.com/docview/1115964583?accountid=45049. Zhang, Jin-yu, Yong Li, and Zhu-ming Chen. 2013. Unit Root Hypothesis in the Presence of Stochastic Volatility, a Bayesian Analysis. Computational Economics 41 (1): 89-100. doi:http://dx.doi.org/10.1007/s10614-012-9319-x. http://search.proquest.com/docview/1270351423?accountid=45049. Antoniou, Antonios & Holmes, Phil (1995). Futures trading, information and spot price volatility: evidence for the FTSE-100 Stock Index Futures contract using GARCH. Journal of Banking and Finance 19: 117-129. [1] Levenbach, Hans. 1982. Time Series Forecasting using Robust Regression. Journal of Forecasting (Pre-1986) 1 (3): 241. [2] Storti, Giuseppe. 2008. Modelling Asymmetric Volatility Dynamics by Multivariate BL- GARCH Models. Statistical Methods & Applications 17 (2): 251-274 [3] Kolade, Sunday Adesina. 2013. Modelling Stock Market Return Volatility: GARCH Evidence from Nigerian Stock Exchange. International Journal of Financial Management 3 (3): 37-46 [4] Rogers, Nina, Kimberly Winson-Geideman, and Imre Karafiath. The Impact of Trading Volume on Reit Volatility Using The Garch Model. Journal of Real Estate Portfolio Management 20, no. 3 (2014): 167-177 [5] F?ªss, Roland, Dieter G. Kaiser, and Zeno Adams. 2007. Value at Risk, GARCH Modelling and the Forecasting of Hedge Fund Return Volatility. Journal of Derivatives & Hedge Funds 13 (1): 2-25 [6] Antoniou, Antonios & Holmes, Phil (1995). Futures trading, information and spot price volatility: evidence for the FTSE-100 Stock Index Futures contract using GARCH. Journal of Banking and Finance 19: 117-129. [7] Liu, Wei and Bruce Morley. 2009. Volatility Forecasting in the Hang Seng Index using the GARCH Approach. Asia ? Pacific Financial Markets 16 (1): 51-63 [8] Daouk, Hazem and Jie Qun Guo. 2004. Switching Asymmetric Garch and Options on A Volatility Index. The Journal of Futures Markets 24 (3): 251-282 . [9] McMillan, David G. and Alan E. H. Speight. 2004. Daily Volatility Forecasts: Reassessing the Performance of GARCH Models. Journal of Forecasting 23 (6): 449 [10] Bunnag, Tanattrin. 2014. Volatility Analysis of International Tourist Arrival Growth Rates to Thailand using Garch and GJR Model. Journal of Environmental Management & Tourism 5 (1): 71-84 [11] Rodrigues, Paulo M. M. and Antonio Rubia. 2004. On the Small Sample Properties of Dickey Fuller and Maximum Likelihood Unit Root Tests on Discrete-Sampled Short-Term Interest Rates. St. Louis: Federal Reserve Bank of St Louis [12] Zhang, Jin-yu, Yong Li, and Zhu-ming Chen. 2013. Unit Root Hypothesis in the Presence of Stochastic Volatility, a Bayesian Analysis. Computational Economics 41 (1): 89-100 [13] Makhwiting, M. R., M. Lesaoana, and C. Sigauke. 2012. Modelling Volatility and Financial Market Risk of Shares on the Johannesburg Stock Xchange. African Journal of Business Management 6 (27): 8065-8070 [14] Wai-Ching Poon, Chee-Keong Choong, and Shah Habibullah Muzafar. 2005. Exchange Rate Volatility and Exports for Selected East Asian Countries: Evidence from Error Correction Model. ASEAN Economic Bulletin 22 (2): 144-159 [15] Kolade, Sunday Adesina. 2013. Modelling Stock Market Return Volatility: GARCH Evidence from Nigerian Stock Exchange. International Journal of Financial Management 3 (3): 37-46 [16] Antoniou, Antonios & Holmes, Phil (1995). Futures trading, information and spot price volatility: evidence for the FTSE-100 Stock Index Futures contract using GARCH. Journal of Banking and Finance 19: 117-129. [17] Antoniou, Antonios & Holmes, Phil (1995). Futures trading, information and spot price volatility: evidence for the FTSE-100 Stock Index Futures contract using GARCH. Journal of Banking and Finance 19: 117-129. [18] Wen-Hsiu Kuo, Hsinan Hsu, and Chwan-Yi Chiang. 2005. Price Volatility, Trading Activity and Market Depth: Evidence from Taiwan and Singapore Taiwan Stock Index Futures Markets. Asia Pacific Management Review 10 (2). [19] Storti, Giuseppe. 2008. Modelling Asymmetric Volatility Dynamics by Multivariate BL- GARCH Models. Statistical Methods & Applications 17 (2): 251-274. [20] Fong, W. M. and K. H. See. 2001. Modelling the Conditional Volatility of Commodity Index Futures as a Regime Switching Process. Journal of Applied Econometrics 16 (2): 133- 163 [21] Bailer, Heiko Manfred. 2005. Robust Estimation of Factor Models in Finance. Order No. 3183337, University of Washington. [22] Grzywacz, Piotr and Krzysztof WoIyniec. 2011. Multi-Scale Volatility In Commodity Markets. Energy Risk, 08, 44-49. [23] Shaikh, Imlak and Puja Padhi. 2014. Stylized Patterns of Implied Volatility in India: A Case Study of NSE Nifty Options. Journal of Indian Business Research 6 (3): 231. [24] Conrad, Christian and Menelaos Karanasos. 2010. Negative Volatility Spillovers In The Unrestricted Eccc-Garch Model. Econometric Theory 26 (3): 838- 862 [25] Wen-Hsiu Kuo, Hsinan Hsu, and Chwan-Yi Chiang. 2005. Price Volatility, Trading Activity and Market Depth: Evidence from Taiwan and Singapore Taiwan Stock Index Futures Markets. Asia Pacific Management Review 10 (2). [26] Conrad, Christian and Menelaos Karanasos. 2010. Negative Volatility Spillovers In The Unrestricted Eccc-Garch Model. Econometric Theory 26 (3): 838- 862 [27] Wen-Hsiu Kuo, Hsinan Hsu, and Chwan-Yi Chiang. 2005. Price Volatility, Trading Activity and Market Depth: Evidence from Taiwan and Singapore Taiwan Stock Index Futures Markets. Asia Pacific Management Review 10 (2).