Eduardo M. A. M.Mendes
2011-Aug-15 22:00 UTC
[R] Regression - how to deal with past values?
Dear R-users I need to fit a nonlinear model to a piece of data. The model to be fitted uses past values of the input and the ouput - something like y(k) ~ f(y(k-1),y(k-2),u(k),u(k-1) ....) (k is time index). As far as I know I could use earth(MARS), nnet and etc but I am not sure how to deal with the past values since most, if not all, examples I saw formula does not take in account past values of any input (let alone output). Is there any package available that deals with this type of regression? (ar (linear) is not what I am looking for). Is there any example out there that I can follow? I have searched the web using words such as fitting and past values but I could not find what I want. Many thanks Ed [[alternative HTML version deleted]]
This may not be helpful, but this sounds difficult enough that you should work with a local statistician rather than trying to get remote consulting here. -- Bert On Mon, Aug 15, 2011 at 3:00 PM, Eduardo M. A. M.Mendes <emammendes at gmail.com> wrote:> Dear R-users > > > > I need to fit a nonlinear model to a piece of data. The model to be fitted > uses past values of the input and the ouput - something like > > y(k) ~ f(y(k-1),y(k-2),u(k),u(k-1) ....) (k is time index). ?As far as I > know I could use earth(MARS), nnet and etc but I am not sure how to deal > with the past values since most, if not all, examples I saw formula does not > take in account past values of any input (let alone output). > > > > Is there any package available that deals with this type of regression? ?(ar > (linear) is not what I am looking for). > > > > Is there any example out there that I can follow? ?I have searched the web > using words such as fitting and past values but I could not find what I > want. > > > > Many thanks > > > > Ed > > > > > ? ? ? ?[[alternative HTML version deleted]] > > ______________________________________________ > R-help at r-project.org mailing list > https://stat.ethz.ch/mailman/listinfo/r-help > PLEASE do read the posting guide http://www.R-project.org/posting-guide.html > and provide commented, minimal, self-contained, reproducible code. >-- "Men by nature long to get on to the ultimate truths, and will often be impatient with elementary studies or fight shy of them. If it were possible to reach the ultimate truths without the elementary studies usually prefixed to them, these would not be preparatory studies but superfluous diversions." -- Maimonides (1135-1204) Bert Gunter Genentech Nonclinical Biostatistics 467-7374
Eduardo M. A. M.Mendes
2011-Aug-16 00:54 UTC
[R] Regression - how to deal with past values?
Many thanks but I am not looking for a remote consulting. Actually it is not difficult but unfortunately I am a newbie as far as writing functions on R. Can I gather from your email that there is nothing available on R that deals with dynamic models (k-step ahead and free-run)? Cheers Ed -----Original Message----- From: Bert Gunter [mailto:gunter.berton at gene.com] Sent: Monday, August 15, 2011 7:42 PM To: Eduardo M. A. M.Mendes Cc: r-help at r-project.org Subject: Re: [R] Regression - how to deal with past values? This may not be helpful, but this sounds difficult enough that you should work with a local statistician rather than trying to get remote consulting here. -- Bert On Mon, Aug 15, 2011 at 3:00 PM, Eduardo M. A. M.Mendes <emammendes at gmail.com> wrote:> Dear R-users > > > > I need to fit a nonlinear model to a piece of data. The model to be > fitted uses past values of the input and the ouput - something like > > y(k) ~ f(y(k-1),y(k-2),u(k),u(k-1) ....) (k is time index). ?As far as > I know I could use earth(MARS), nnet and etc but I am not sure how to > deal with the past values since most, if not all, examples I saw > formula does not take in account past values of any input (let aloneoutput).> > > > Is there any package available that deals with this type of > regression? ?(ar > (linear) is not what I am looking for). > > > > Is there any example out there that I can follow? ?I have searched the > web using words such as fitting and past values but I could not find > what I want. > > > > Many thanks > > > > Ed > > > > > ? ? ? ?[[alternative HTML version deleted]] > > ______________________________________________ > R-help at r-project.org mailing list > https://stat.ethz.ch/mailman/listinfo/r-help > PLEASE do read the posting guide > http://www.R-project.org/posting-guide.html > and provide commented, minimal, self-contained, reproducible code. >-- "Men by nature long to get on to the ultimate truths, and will often be impatient with elementary studies or fight shy of them. If it were possible to reach the ultimate truths without the elementary studies usually prefixed to them, these would not be preparatory studies but superfluous diversions." -- Maimonides (1135-1204) Bert Gunter Genentech Nonclinical Biostatistics 467-7374