Model A)

a) Carry out the required model estimations, hypothesis testing and model verifications as required by consulting chapters 2, 3, 4, 5, and 6 of the text book.

b) Provide your answers in a word file with short explanations with results from your Eviews estimation inserted as appropriate. You need place your EVIEWS and WORD files into a folder (folder name should be your lastname_firstname), zip or rar that folder and upload the compressed file to Blackboard.

Variable Creation:

Using Eviews create the following variables:

i) Daily log returns (in percentage) for Caterpillar Inc. (CAT), and S&P 500 index (SPX) using PX LAST.

ii) Name return variables as R_CAT and R_SPX

You need to submit your Eviews files with data/regression output tables/equation estimates (frozen as tables) and a Word file explaining your findings with Eviews results (tables/outputs from Eviews copied/pasted into Word as picture).

Statistical Modeling and Testing:

For all the questions below, you need to refer to specific estimated coefficients, t or F statistics, p or R2 values, DW, AIC, SBIC, or HQ values.

Model A) Using the risk-free-rate create excess return variables and estimate the Beta coefficient for CAT as described in the textbook under CAPM model. Name excess return variables as ER CAT and

ER SPX.

a. What is the slope and intercept from your estimation? What do they indicate?

b. Are intercept and slope coefficients statistically significant (explain your hypothesis tests)?

c. Can the Beta of this stock (CAT) statistically equal to 1.5?

d. What is the R2 of the CAPM regression? What does it indicate?

e. Is the model you estimated statistically significant?

f. Does the C:APM model with the samp!e data you used violate the assumptions of CLRM ?, i.e. is there heteroscedasticity in the estimation results? Is there autocorrelation in the estimation results? (Provide statistical evidence/test results/graphs etc. to support your answers).

Model B)

Perform univariate time-series analysis of BOTH the daily log returns for the stock (CAT) and the daily log returns for the S&P 500 index (SPX), using the R_CAT and R_SPX variables (don't use ER_C:AT or ER SPX). The analysis should include (as discussed in Chapter 6).

a. An examination of the autocorrelation and partial autocorrelation functions (for both C:AT and SPX )

b. An estimation of the information criteria for each ARMA model order from (0,0) to (5,5) for both CAT and SPX. {For this step, you HAVE to use the Eviews program "eviews best arma_E9.prg" posted in Blackboard. You need to make necessary modifications for variable names in the program]

c. An estimation of the models for both CAT and SPX that you feel most appropriate given the results that you found in part b (which is the best ARMA(p,q) model?).

For parts a, b, and c, you need to use data from beginning of Jan 2005 to the end of Dec 2014, create information criterion tables for AICI, SBIC, and HQ. NOT the entire sample like the examples in the book! {All these steps need to be repeated for the stock CAT and the S&P 500 index (SPX)}.

d. The construction of a forecasting framework to compare the forecasting accuracy of 1. Your chosen (best) ARMA model

ii. An arbitrary ARMA(1,1)

For part d, forecast period is from the beginning of Jan 2015 to the end of Dec 2015. You need to test dynamic as well as static forecast of models i) best ARMA and ii} ARMA(1,1) for both the stock CAT and .S&P 500 index (SPX).

e. Based on your answers to above questions, which univariate time-series model best describes (1) returns of the stock CAT, (2) returns of S&P 500 index (SPX)?

f. Which univariate model provides the best forecast accuracy? (Use MSE, MAE, Theil, etc. measures to make your comparison-as explained in Chapter 6).

Attachment:- HW.rar