Reference no: EM131962413

A recent study by Ray Fair at Yale University analyzed the relation between stock prices and risk aversion in the US stock market over time. Fair examined the performance of sixty-five companies consistently included in the S&P 500 hypothesizing that stock prices would be a positive function of both earnings (EARN) and dividend growth (DIV) but negatively related to the degree of risk as measured by a company’s Beta (BETA), where a higher Beta indicates higher risk. Fair estimated an equation of the following form using the STOCKS Stata dataset posted on Canvas: a) What is the functional form of the equation and do you think it is an appropriate choice? Why or Why not? Functional form is the relationship between a dependent variable Y and Independent variables X1,X2,X3 Regarding with this one, I think the Beta *β1 should be negative since it is the negatively related to stock prices. b) Construct scatterplots between the dependent variable and each of the independent variables. c) Calculate descriptive statistics for each of the variables and place your results in a table showing the mean, median, variance, standard deviation, skewness, kurtosis, and coefficient of variation. d) Calculate a pairwise correlation matrix between the variables. e) Estimate the model with all the variables included assuming Homoscedasticity and then correcting for the possible presence of Heteroscedasticity and store your results for each model. f) Do you notice any differences in the estimations? What are they? g) State and test the model hypotheses (individually and for the entire model) at the .05 level of significance and report your results for each specification. h) Does one of the variables appear to be an irrelevant variable? Why or Why not? i) Conduct a Ramsey RESET test and report your results. j) Does the specification test indicate that any of the models are misspecified? k) Now reestimate your model dropping your potentially irrelevant variable again under the assumption of Homoscedasticity and correcting for the possible presence of Heteroscedasticity and store your results. l) What is the problem with dropping variables from a model based solely on statistical significance? m) Construct a table of your four regressions reporting the number of observations and goodness of fit measures: r2, the adjusted r2, the AIC, and the SIC. n) Which model do you prefer? Why?