motivated by email from Darren S..
Okay, here's what we want to do:
 Pick some stock or fund or index.
 Download five years worth of daily prices.
 Calculate (about 1200) daily returns.
 Determine the Compound Daily Returns for Mondays, Tuesdays, etc.
 Calculate the Compound Daily Returns for January, February, etc.
 Calculate what percentage of the time the return was positive.
 Plot a chart and ...
>Wait! Compound Daily Returns?
Yeah. We look at the returns for Mondays only. Suppose they're r_{1}, r_{2}, ... r_{n}.
Then we calculate (1+R)^{n} = (1+r_{1})(1+r_{2})...(1+r_{n}), the cumulative Gain Factor, for Mondays.
(That'd be what happens to $1.00 after 5 years.)
Then we find R, the Compound Daily Return ... for Mondays.
Then we do this ...
>For Tuesdays and Wednesdays. Yeah, I get it.
And for all the months of January, then February, then March , then ...
>Why Compound Daily Returns? Why not Average Daily Returns
Yeah, we can do that, too. The Compound Gain Factors will be smaller because of volaitlity.
Anyway, we'd get something like this:
>I assume I just click on the picture to download the spreadsheet.
You got it!
