This preview has intentionally blurred sections. Sign up to view the full version.View Full Document
Unformatted text preview: CORNELL UNIVERSITY STSCI 4550 / ILRST 4550 / ORIE 5550 Applied Time Series Analysis, Spring 2012 Professor David S. Matteson Assignment #6 Due: Monday, April 2 1. Consider the monthly simple return of CRSP Decile 1 portfolio from January 1988 to December 2007. The data are in m-dec1n10.txt (date, Decile-1 return, and Decile-10 return). It is well-documented that there is a January effect for smaller firms. To model this January effect, we can use January dummy variable in a regression setup, resulting in a regression model with time series errors. To do this, use the following command to create January dummy in R for 20 years of data: jan = rep(c(1,rep(0,11)),20) where “rep” means “repeat”, i.e. rep(0,11) means repeating 0 for 11 times. Identify a regression model with time-series errors for the Decile 1 simple returns. Perform model checking using Q(24) on the residuals, and write down the fitted model. Test the null hypothesis that the January effect is indeed significant....
View Full Document
- Spring '12
- Regression Analysis, fitted model, David S. Matteson, CORNELL UNIVERSITY STSCI