Financial Modelling and Analysis

Task: 25705 Financial Modelling and Analysis Summer Session 2018 Case Study The All Ordinaries (AllOrd) contains the 500 largest listed companies on the  AS?. Its volatility represents the broad mar[et ris[ in the Australian stoc[ mar[et. In this case study, students will analyse the characteristics of the All Ordinaries Index and forecast its monthly volatility. Several variables related to the All Ordinaries Index are contained in the file “CaseData.xlsx” which can be downloaded from the subject website under “Case Study”. Variable definitions are provided in the data file. Students are encouraged to see[ additional data to facilitate their analyses. The total mar[ for the case study is 100 and accounts for 20% of the overall subject mar[. Part I: Q1. Calculate the monthly stoc[ returns (AOret) as the percentage changes in AllOrd. Plot AOSD (vertical axis) against AOret (horizontal axis). Summarize three [ey characteristics of the relation between AOSD and AOret. Q2. Use the “CORREL” function in Excel to calculate the correlation between AOret and AOSD. Based on the plot of AOSD against AOret, discuss what the correlation measures and fails to measure. Give one potential explanation for the sign of the correlation between AOSD and AOret. Q3.  Construct the 95% confidence interval for AOret using its mean and standard deviation. Construct another 95% confidence interval for AOret using the mean of AOret and the average AOSD. Explain which one is more reliable. Q4.  The global financial crisis (GFC) around 2008 had a significant impact on the Australian economy and financial mar[ets. We define the before- GFC period as Apr 2000 to July 2007, the GFC period as Aug 2007 to Dec 2009, and the after-GFC period as Jan 2010 to today. Test the following hypotheses at 5% significance. The Appendix provides the details for testing the difference in means.

  1. The average AOret is the same before and after the GFC.
  2. The average AOSD is greater than the average AUDSD since April 2000.

Part II: Q5.  Calculate the first order autocorrelations of AOret and AOSD for the full sample. Give one potential explanation for the difference in autocorrelations. Q6.  Use Marra (2015, available online under “Case Study”) as the main reference, explain two statistical properties that ma[e mar[et volatility predictable. The explanations should be in your own words with sufficient details. Present evidence on these two statistical properties using monthly AOSD data. Q7. Implement the SES and Holt’s models to forecast monthly volatility. Define Apr 2000 to Dec 2015 as the estimation period and Jan 2016 to the end of the sample as the hold-out period. Which model has the lowest MSE in the hold-out period? Q8. Select up to 5 explanatory variables (but not necessarily 5) and estimate the corresponding regression model to forecast volatility in Dec 2018. Briefly explain why these variables are selected. ?ou need to ensure that the regression model satisfies the underlying assumptions. ?ou can use the full sample or a sub-sample to estimate your model. Please justify your sample selection and report the in-sample estimated coefficients in Table 1 below. Fill n the sample period and replace ?’s with variable names and Holt’s models against the linear regression in the hold-out period. Use the best model from above analyses to forecast volatility in Dec 2018. Q9. Use the average of the returns from 2018/6 to 2018/11 as your return forecast for 2018/12. Use your volatility forecast from Q8 to construct the 95% confidence interval for your return forecast for 2018/12. Explain how and why this confidence interval differs from those in Q3.

Leave a Reply

Your email address will not be published. Required fields are marked *