Question

# If you are not sure about something for this assignment, you can state your assumptions. Excel Data attached. For

questions 1-3 use *monthly data:* Exxon stock price $US, Crude Oil Price $US, and Standard and Poors 500 stock index (S&P500). For Question 4 use *annual data:* S&P500 index and S&P500 P/E ratio.You can use Excel, SPSS, or SAS etc.

**1.** Assume you are an investor who wants to examine the risk level in Exxon, an oil stock For Exxon stock price, compute the following risk measures (you can look up Exxon on internet if you want to find out more about it):

a) compute mean, variance, standard deviation, skewness, kurtosis, coefficient of variation, minimum, maximum Briefly define each risk measure and explain the importance of each for measuring risk. Also test for normality. (You could use the Kolmogorov-Smirnov normality test if you wish. Or, you could try other normality tests such as: Jarque-Bera test, the Anderson-Darling test, the Shapiro-Wilk test, or the Pearson's chi-square test). Plot the frequency distribution (histogram).

**2.**Now take the percentage price changes (__returns__) in Exxon stock price.

Note e.g. for monthly data, %price change=[log (this months price)- log(last month price)]*100

a) compute mean, variance, standard deviation, skewness, kurtosis, coefficient of variation, minimum, maximum. Also test for normality. Plot the frequency distribution.

b)briefly compare your answers here to 1. above and explain any important differences, and why they may occur.

**3.** Using __returns__, e.g. percentage price changes as in 2. use regression to compute

a) the Beta of the stock

Exxon stock price = f(S&P500)

Interpret the Beta coefficient, what type of risk does it measure? What does this tell the investor?

b) the Beta

Exxon stock price = f(Crude Oil Price)

Interpret this coefficient, what does this tell the investor?

c) Test for autocorrelation in equation a) and b) using DW statistic, is there autocorrelation? Would you expect to find autocorrelation?

d) Would you expect the coefficient in a) or b) to be close to a value of one? Why or Why not?