tag:blogger.com,1999:blog-107568321062020427.post237779198559676937..comments2023-09-19T00:48:06.643-07:00Comments on Intelligent Trading: Practical Implementation of Neural Network based time series (stock) prediction -PART 4Intelligent Tradinghttp://www.blogger.com/profile/17765336450326139518noreply@blogger.comBlogger2125tag:blogger.com,1999:blog-107568321062020427.post-45763875335718394802012-01-28T14:24:02.553-08:002012-01-28T14:24:02.553-08:00Kleyson,
I don't have a simple answer I'm...Kleyson,<br /><br />I don't have a simple answer I'm afraid. But there are many different things to try; such as pre-processing data (scaling/normalizing), transforming, and or smoothing first.<br /><br />ITIntelligent Tradinghttps://www.blogger.com/profile/17765336450326139518noreply@blogger.comtag:blogger.com,1999:blog-107568321062020427.post-47068731855318723462012-01-26T09:38:05.956-08:002012-01-26T09:38:05.956-08:00Hi,
May you give more details about who to do the...Hi,<br /><br />May you give more details about who to do the first and second de-trend ?<br /><br />I am trying to use NN to predict a time series, not financial, so I do 2 differencing of the serie and after I apply a log function. But the result is bad.<br /><br />Serie Diff1 Diff2<br />3 <br />5 5-3=2<br />6 6-5=1 1-2=-1<br />3 3-6=-3 -3-1=-4<br />4 4-3=1 1-(-3)=4<br /><br />Maybe your process can give a better result.<br /><br />Regards.Kleyson Riosnoreply@blogger.com