[R-SIG-Finance] Eman NN using RSNNS package

[Joshua Ulrich]

Please do not cross-post, or at least tell people that you are
cross-posting and link to the other places you've posted:
http://stackoverflow.com/questions/30802910/out-of-sample-prediction-using-elman-in-rsnns-package
If someone does not follow all the sites you post to, they could spend
time answering your question that has already been answered on another
site.

On Fri, Jun 12, 2015 at 6:26 AM, Abhay Bhadani <abhadani at gmail.com> wrote:
> Dear all,
>
> I am trying to use elman function in package RSNNS to train and test a
> neural network model for stock price prediction.
>
> Based on my limited understanding from the manual and a research article on
> RSNNS package, I interpret that the network can be trained using
>
> model <- elman(patterns$inputsTrain, patterns$targetsTrain,
> size = c(8, 8), learnFuncParams = c(0.1), maxit = 1000,
> inputsTest = patterns$inputsTest, targetsTest = patterns$targetsTest,
> linOut = FALSE)
>
> and the values for the new dataset can be obtained using prediction <-
> predict(model, newdataset).
>
> I am trying predict the future stock price using its past value as input (I
> am considering the univariate case). How do I go about the case of
> out-of-sample prediction? For instance, I have 400 data points. Suppose I
> train and test the model using 300 and 100 data points, respectively. Now
> how do I predict the 401th data point using elman function in RSNNS
> package?
>
> Your suggestions are much appreciated!
>
> Thanks!
>
>         [[alternative HTML version deleted]]
>
> _______________________________________________
> R-SIG-Finance at r-project.org mailing list
> https://stat.ethz.ch/mailman/listinfo/r-sig-finance
> -- Subscriber-posting only. If you want to post, subscribe first.
> -- Also note that this is not the r-help list where general R questions should go.



-- 
Joshua Ulrich  |  about.me/joshuaulrich
FOSS Trading  |  www.fosstrading.com