Forecasting Crude Oil (Spot Price) Volatility Dissertation

Forecasting Crude Oil (Spot Price) Volatility - Dissertation illustrationDaily prices for crude oils atomic number 18 effective in volatility forecasting. 17 It was also imperative to use the two cluster analysis in the paper. 17 In the case of GARCH to obtain the unknowns the formula was applied where the initial value Xk was taken to be 25.56 where a= 0.001 (fixed) 17 b= 0.00 18 c= 0.00 18 In using the same formula the values for a, b and c were P-GARCH established to be 18 a= 0.001(fixed) 18 b= 0.394 18 c= 0.050 18 Xk= 25.56 18 For GARCH GJR, the values were found to be 18 a=0.001 (fixed) 18 b= 0.488 18 c= 0.110 18 Xk= 25.56 18 for E GARCH a=0.001 (fixed) 18 b= 0.488 18 c= 0.11 18 From the findings captured in the spread sheet, we can derive various important factors about the GARCH family models and answer important questions arising from the same. These are 19 The data should be indoors range in order to get rid of outlier values.The data is reliable since the projection/ fo recasted values are indoors limit. There are no outlier values as a result of projection. 19 The null hypothesis Garch models predict uniformly 19 Alternative hypothesis- GARCH models predictions differ. establish on the results, it is clear that there exists variants among the four models. Thus it is rational to conclude that the alternative hypothesis holds. 19 The best model should be as nestled to the baseline as possible. GARCH is a replica of the baseline and hence cannot be taken to be the best.Of the four GJR GARCH varies the least from the baseline hence is the best. 19 EGARCH has the largest variation from the baseline hence is the worst. 19 20 BIBLIOGRAPHY 20 APPENDICES.20 METHODOLOGY AND DATA Introduction Volatility is a concept that refers to the spread of all possible outcomes within an uncertain variable. Finance discipline has various uncertain variables such as prices of products, returns on assets, and share prices amongst others (Olowe, 2010). Modeling and for ecasting of volatility have been attributed to increasing doubt in financial aspects and components (Day & Lewis, 1993). Oil price fluctuations in the global arena experience operative uncertainties thereby invoking interests amongst financial and market participants (Kang, Kang, & Yoon, 2009). The main reasons explaining such significant interest include the fact that oil price fluctuations reckon decision making process for both producers and consumers in addition to investors decisions. Whereas oil price fluctuations affect strategic planning and appraisal of projects for producers and consumers, investors continue to face challenges in investment, allocation of portfolios, and management of risks decisions (Campbell, Lo, & MacKinlay, 1997). Policy and decision making within the oil markets require correct forecasting of the crude oil prices (Olowe, 2010). Attaining accurate and adequate forecasting require adequate and accurate data. In most cases, daily prices of crude oil are used to predict or forecast volatility (uncertainty) for purposes of developing effective policies and decision making processes (Campbell, Lo, & MacKinlay, 1997). Forecasting volatility of crude oil prices have been through for a