MDA - Market Direction Analysis

[MechanicalMethod]

My earlier description neglected to consider unchanged vol. Here's an adjusted model.

up = 60
dn = 30
unchanged = 10

total vol = 100

That's actual raw volume rather than percentage volume. Although it totals 100 it's only by design to facillate finding logic errors.

PNPV = positive volume / (positive v + negative v + unchanged v)


PNPV = 60 / (60+30+10) = .6 or 60% so PNPV looks good to me. But I don't think it's Percentage "NET POSITIVE V". I thought NPV would be:

net / (positive v + negative v + unchanged v);
where net = positive v - negative v;

I might be misunderstanding the objective and thus there's a reason you're using positive v instead of net v. I don't know because I don't have a feel for where this is heading yet.

PVO = (13-day SMA - 21-day SMA)/ 13-day SMA)

That part is really cool because it begins with MACD of the signal and divides by the leading half of the signal. I'll have to pick this up at another time because it's too much to think about right now.

Thanks for the study. I'm anxious to see what develops out of it. A guy I really admire always says he didn't know enough to know that it couldn't be done so he just kept trying it a different way. He's quite successful. The other thing I've picked up from him is when he describes associates he's had that were extremely successful he always mentions how bent their thinking process was from the norm. just fwiw, it might not mean anything to anyone on this thread but it's my opinion from studying successful traders that scientific technique only goes so far and without the drive to create something unusual it doesn't matter how many stats classes one takes.