correct sign 52.58
most valuable input CLOSE and lag(1)
most valuable input CLOSE and lag(1)
Optimized Risk vs Reward Equation 335 replies
Loss Less Optimized Strategy 63 replies
Optimized Parameters in Metatrader Testing? 3 replies
"Optimized Parameter" feature in EA? 0 replies
Dislikedhi madcow,
i work in amibroker & mt4 language is new to me.i wanted to convert igorads nonlag ma into afl but cant.
pls can u give the afl for igorads nonlag ma & also the ULLMA in afl.
thnks for ur help.Ignored
DislikedConclusions
Correct sign varied between 51.6% and 62%, its very low. Variation depend from number of bars and its clear impact of Law of Large Number ==> Predictive power of this setup will be very low because even for in sample period was not possible to get high correct sign value. It was confirmed by out of sample period test.
Valuable inputs were chosen sometimes filters and sometimes as Lag/Momentum. There is no clear advantage of using
Filters. For reference signal sin + cos filters were not chosen at all !!!
Fisherized price was never...Ignored
DislikedI am sorry for the delayed response. I have not been monitoring this thread for some time. I don't recall having Igorads nlma in AFL, but here is the afl code for the ULLMA as well as almost every filter known to man. (You must change the extension to afl. it is not a zip file, but FF requires certain extensions). Just apply ULLMA1(price) to one of the HA functions. If you have questions, PM me. This is not an afl forum.
Attachment
MadCowIgnored
DislikedI don't follow climatology at all, but my ears instantly perk up when you mention it in this forum.
We're dealing with almost exactly the same problems. Predicting the (short term) future of a non-linear dynamic system. No doubt they have some techniques that would be useful.Ignored