Instructions: Type out your work in a readable form using a document editor of your choice.
Also attach the Matlab code of your work to the submission on Canvas. Code should be easy-to
read and should be able to run from my work station without any difficulty. Do not include any
irrelevant code or texts. Set rng(‘default’) for Matlab simulation so that I may check your work.
1. The sampling period is important in empirical applications as the results are sensitive to it.
For the manager sentiment application, we used 144 monthly observations over the period
2003:01-2014:12 in class. Now consider a shorter period, 2005:01-2014:12 using the same
dataset sentiment data.xlsx. Run the predictive regression(horizon 1) of excess return
on the manager sentiment index AND the 1st principal component of 12 macro-economic
variables(14 minus two redundant ones).
a. Report the coefficient of the manager sentiment index and its t-statistic.
b. Report the coeffcient of for the 1st PC from monthly GWData.xlsx using this approach
and its t-statistic.
c. Report the 12 coeffcient estimates for the 12 macroeconomic variables from monthly GWData.xlsx
using this approach(with the 1st PC). You should have 12 numbers.
d. Answer (a), (b), and (c) again using the first four principal components of 12 macroe
conomic variables from monthly GWData.xlsx.