FIN3018 Financial Econometrics and Data Science
Hello, dear friend, you can consult us at any time if you have any questions, add WeChat: daixieit
FIN3018
Financial Econometrics and Data Science
Tutorial Week 6
October 2022
Download macro_ 1.csv from Canvas. Variables in this dataset:
Variable Name |
Description |
er_msoft |
Monthly excess returns of the Microsoft stock |
er_sandp |
Monthly excess returns of the S&P 500 index |
r_term |
Monthly excess returns of ten years Treasury bill |
d_money |
Monthly difference of a measure of narrow money supply |
d_prod |
Monthly difference of an industrial production index |
d_credit |
Monthly difference of a consumer credit series |
d_inflation |
Monthly difference of inflation |
The question we want to discuss is: what are the factors influence the excess returns of the Microsoft stock?
Q1. Estimate a regression with monthly excess returns of the Microsoft stock as a dependent and all other variables as independent variables.
a) Print the result of regression
b) What is the value of R-squared of this model? What is the value of adjusted R- squared?
c) Write down H0 and H1 of F test in this regression result. Do we reject H0?
d) Estimate another regression with the same dependent variable and independent variables are: er_sandp, r_term, d_inflation
Q2. Perform the following partial F test based on the regression we run in Q1.a:
a) d_money=0 and d_prod=0 and d_credit=0"
b) r_term=0
Q3. a) Applying forward variable screening method, write down the final model
b) Applying backward variable screening method, write down the final model
We now change the question we focus into: Do difference of a measure of narrow money supply influence excess returns of the Microsoft stock? So, monthly difference of a measure of narrow money supply is the independent variable we discuss, other independent variables are control variables.
c) Applying forward variable screening method, write down the final model
d) Applying backward variable screening method, write down the final model. Please interpret the final model as well.
Reference code:
install.packages("MASS") ; library(MASS);
step(TheModelStart, direction="forward") step(TheModelStart, direction="backward")
scope=list(lower=BaseModel,
scope=list(lower=BaseModel,
upper=FullModel),
upper=FullModel),
2022-10-27