Which index should one use to produce excess returns.
Can we regress monthly but on daily return. The returns will be the same however the OLS estimate of $\hat{\beta}$ will not, since this is obtained as the sum of squared of residuals (hence quadratic).
Shall we use closing prices or adjusted closing prices
Shall we use return as diff or as procentage
Shall we take t-1 two times or only one to secure no look ahead bias.
When calculting the exponential weighted average and annulize it one get some abnormes high values. In the calculation we have used increasing data and the weights is not converging at only a subset of the data. However should we do something different?
P. 236 column 2: "The sharpe ratio is statistically different from zero", how do they test this?