I am working on a piece of work but I am having a bit of trouble understanding exactly what it is the lecturer wants us to do.
The piece of work is centred around building forecasts and regression of a company's monthly return (the dependent variable) using the VIX, term spread, SP500 return, and CPI as the dependent variables.
"Based on the above regression model check if the average return of your firm
(controlling for the effect of the four economic variables) is different in the period after January 2009 (inclusive) compared to the period before January 2009."
I have answered this question by looking at the mean return produced by the original regression model for before and after Jan 2009. I've then conducted a two tailed t-test to compare the means. Where:
H0: means are equal
HA: means are not equal
Does this seem like I am on the right track? I am confused by the "controlling for the four other economic variables". We haven't covered two-tailed t-tests to compare means in this module so I feel as if I may have gone astray. I think possible I should create two separate regressions (i.e. using the data from before Jan 09 and after Jan 09). Thoughts?
"Suppose you estimate a regression of the return of your firm on the return of the S&P 500 and the square of the return on the S&P 500. What is the meaning of the coefficient of the squared term? Note: This is a discussion question, so no additional analysis is required"
My main points:
- beta (coefficient = 1.95) represents the change in the dependent variable in response to a unit change in the independent (sp500 squared);
- a negative unit change in sp500 would result in a positive unit change in sp500 squared ==> a positive, or negative, unit change in the sp500 would, therefore, on average, produce a 1.95 change in companies return.
- squaring makes all values positive
- p-value > 0.05 of each parameter.
- F significance > 0.05 of entire model.
I feel like I have a relatively good grasp of the course material it's just the above two questions weren't really covered in the lectures/labs.
Thanks in advance.
1 - before Jan 09
0 - after Jan 09.