the formulas that divvy up the days volume as "buying volume" versus "selling volume" are the ones with "accumulation/distribution" or "money flow" in the names. There's a few different ways to view this, but most use (h-l) in the denominator, and some combination of o-h-l-c in the numerator. Chaikin uses {[(c-l)-(h-c)]/(h-l]}*V; several people use [(c-o)/(h-l)]*v.
I kind of figured that Money Flow, Accum/Distr, and the like do this sort of thing, yet in Wow Pro, they do not show the formulation of the calculation for which I feel more comfortable with. Somehow, I just like knowing how it is calculated so that I can change things if I am not happy with their traditional formula.
By all means, experiment! You may find something very useful--even if it's not what you originally set out to find. I'm looking at some Accum/dist experiments also, so we'll have to compare notes. Mine are real obscure though, so I want to mess with them for a while before I post. They make more logical sense to me though, so I may use them in place of the "conventional" a/d or money flow indicators even if they are just equal, on average, in their accuracy.
I am thinking along the same lines. Although whatever formula I come up with might, or might not do what I want, it is possible that the formula might have merit for some use instead of the standard canned indicators. Who knows what I will find, but it is interesting and keeps my mind occupied so I have fun in doing it. Good luck with yours and post back when finished with it, I will do the same.
I was interested in your VLC calculations and tried to duplicate what you were trying to post, yet I did not see how you were using it as an evaluation/plot. Care to send me your excel spreadsheet on this so I can see the formulas and what they are trying to do. Specifically, once plotted, what are you looking for to happen? Close to be out of the range on the upside?
Tim
Now this was kind of neat. Was able to come back and edit this message. Was just reading other forums and wondered if it might not be a prudent idea to disregard a certain amount of volume in a calculation if the volume was such that it did not move price a certain percentage? Example: Volume is 100,000 a day average and past couple of days have been 200,000 and price moves up normally, yet maybe the 3rd day the volume spikes to 900,000 and no appreciable price increase. I might add this into my calculations that if volume is more than 5 times normal average volume, then cut the volume off at that amount so the plot will not be skewed with spikes that really didn't do anything. What do you think about this? I will play with it.
Tim |
