I have a portfolio of about 50 stocks from which I have picked about 4-8 stocks each year to see whether they perform better than the 50 stocks as a whole. Is a paired t-test a good way to test whether there is a difference in the average returns of these two portfolios for over for example 15 years?
Secondly, I want to test whether the spread, i.e. the 'additional return', earned by these 5 stocks compared to the portfolio of 50 stocks change over time, e.g. if it is higher between 2000 and 2004 than between 2005 and 2009. Could anyone think of a good way of testing this? Would it be suitable to pool the data from 2000 to 2004 and 2005 to 2009, for example?
Reply With Quote