I'm probably being a little too finicky at this early stage but I tripped over the issue and thought I would see if anyone else had factored this in - obviously you have.
I should say that I am not using Excell and not using any commercial TA software for this. I'm writing my own stuff - It's not all that hard and is much easier to mess with than commercial software (for my purposes anyway). I'm not doing anything overly fancy indicator wise - mainly looking at Standard Deviations and ATRs at this early stage.
My first step is just getting the historical data up the way I want it. As long as I can get a file of historical dividends and dates I can easily merge them in to the historical prices.
This is where my original question comes in. My feeling is that the expected dividend could be considered to be factored into the price over the "earning period" (eg six months). Obviously, early in the the piece the effect of the dividend would be miniscule, building up to a "crescendo" the day before ex. The method that you mentioned would give me a problem as I would get a "lump" on dividend announcement day.
Thinking on it, this same effect would have to be considered in option pricing. I wonder what formula they use?
Of course the other way to deal with it is to look at FX instead (no dividends). Just not sure where to get historical FX prices.
