Reference no: EM132578871

MFE 6220 Quantitative Analysis in Equity Markets - Ohio University

Question 1

In this question we investigate the effect of wage changes on the inflation rate. Such effects can be from the demand side or the supply side. On the supply side, we expect wage increases to increase cost of production and to drive up prices. On the demand side, wage increases mean greater disposable income, and a greater demand for goods and services that also pushes up prices. Irrespective of the line of reasoning, the relationship between wage changes and inflation is likely to be a dynamic one; it takes time for wage changes to impact inflation. To investigate this dynamic relationship, we use quarterly data on U.S inflation (INF) and wage growth (WGWTH) from 1970Q2 to 2010Q1. These data can be found in the file namedinflation.

Estimate the model INF_t= δ+δ₀ WGWTH_t+v_t. Show the estimation output and check for serial correlation in the errors of the model.

Now estimate the following model
INF_t= δ+θ₁ INF_t-1+θ₃ INF_(t-3) +δ₀ WGWTH_t +v_t + δ₁WGWTH_(t-1) . Show the estimation output and check for serial correlation in the errors.

For the two estimations, obtain inflation forecasts for the last six quarters of the sample.

For each case, plot in one graph the actual and the forecasted inflation rate(along with the 95% confidence interval band for the forecasts) only for the forecast horizon. Do the forecasts closely resemble the actual data?

Compute the RMSE value for the two models and determine which model provides the best inflation forecasts.

Question 2

Themarket modelin Finance is shown in equation (1) below.

R_t = α + βR_m,t + ε_t (1)

An updated market modelis given in equation (2) below.

R_t = α + β₁R_m,t + β₂R_comp,t + ε_t (2)

In this model, R is the stock return for an individual company (i.e. the company you will select) and R_m is the aggregate market return (i.e. S&P 500 return). R_comp is the stock return for a competitor company in the same industry as the company you selected.

Go to Yahoo Finance and download daily datafor the time period from January 1, 2010 to May 31, 2020 for the following three: a company of your choice, a competitor company for the company of your choice, and the S&P 500 Composite index (^GSPC).

For the companies that you selected, clearly indicate their names and ticker symbols.

Import the Yahoo Finance data into SAS. Then write a SAS program that does all of the following:

a. Merge the three data sets into one SAS dataset.

b. Compute the continuously compounded (CC) stock return (in percentages) for your company, its competitor company, and for the S&P 500 index. Remember to compute the CC returns using the adjusted closing prices. Then obtain the basic summary statistics onlyfor the three returns.

c. Estimate equation (2) and show the regression output. Note that this is the original model and it has ARMA(0,0) terms.

d. Estimate equation (2) including ARMA(1,0) term and show the regression output.

e. Estimate equation (2) including ARMA(1,1) terms and show the regression output.

f. Estimate equation (2) including ARMA (2,1) terms and show the regression output.

For each of the foursmodels,

g. Forecast values for your company's stock return for the last one month of the sample.

h. Plot in one graph the actual and forecasted return (including the 95% confidence interval for the forecasted return) for the forecast horizon (i.e. the last month of the sample).

i. Obtain the RMSE values for each model and determine which model provides the optimum forecasts for your company's return.

j. For the model that you selected in part (i), does the actual return always fall inside the 95% confidence interval for the forecasted return?

Note: Even if the point estimates for the forecasted return deviate from the actual return observations, you will have some confidence in your forecasts if the actual company return does in fact fall within the 95% confidence interval.

For this assignment, use only the PROC ARIMA code.

Attachment:- Homework Assignment.rar