Re: [R] High dimensional optimization in R
Hello, Genetic algorithm can prove handy as well here. see for instance https://cran.r-project.org/web/packages/GA/vignettes/GA.html with non-convex objective functions I usually try a genetic algorithm for a few rounds then finish using nlminb Best regards, Jeremie Marc Girondot via R-help writes: > I fit also model with many variables (>100) and I get good result when > I mix several method iteratively, for example: 500 iterations of > Nelder-Mead followed by 500 iterations of BFGS followed by 500 > iterations of Nelder-Mead followed by 500 iterations of BFGS > etc. until it stabilized. It can take several days. > I use or several rounds of optimx or simply succession of optim. > > Marc > > Le 28/11/2018 à 09:29, Ruben a écrit : >> Hi, >> >> Sarah Goslee (jn reply to Basic optimization question (I'm a >> rookie)): "R is quite good at optimization." >> >> I wonder what is the experience of the R user community with high >> dimensional problems, various objective functions and various >> numerical methods in R. >> >> In my experience with my package CatDyn (which depends on optimx), I >> have fitted nonlinear models with nearly 50 free parameters using >> normal, lognormal, gamma, Poisson and negative binomial exact >> loglikelihoods, and adjusted profile normal and adjusted profile >> lognormal approximate loglikelihoods. >> >> Most numerical methods crash, but CG and spg often, and BFGS, >> bobyqa, newuoa and Nelder-Mead sometimes, do yield good results (all >> numerical gradients less than 1) after 1 day or more running in a >> normal 64 bit PC with Ubuntu 16.04 or Windows 7. >> >> Ruben >> > > __ > R-help@r-project.org mailing list -- To UNSUBSCRIBE and more, see > 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. __ R-help@r-project.org mailing list -- To UNSUBSCRIBE and more, see 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.
Re: [R] Seek help - plm package (Error message: duplicate 'row.names' are not allowed)
Hi david,
The formatting of the data frame looks like the Province and Year
columns have gotten stuck together. This probably has something to do
with your Excel spreadsheet or the function that you are using to read
it in. If there is are fewer column names than columns, this error is
likely to happen. As your data did not get through, this is a guess,
but it might help.
Jim
On Sat, Dec 1, 2018 at 1:15 AM Wong David wrote:
>
> Dear Madam/ Sir,
>
> When I used the 'plm' package and import data into r last week, I found that
> everything was running smooth. However, when I used the 'plm' package today,
> I found the following error message:
>
> > pdata <- pdata.frame(mydata, index=c("Province","Year"))
> > pooling <- plm(Y~X, data=pdata,model="pooling")
> Error in `row.names<-.data.frame`(`*tmp*`, value = c("Anhui-2006",
> "Anhui-2007", :
> duplicate 'row.names' are not allowed
> In addition: Warning message:
> non-unique values when setting 'row.names':
>
> I did not find out the above error message when running panel data regression
> over one year. Moreover, I attempted to manipulate the data many times in the
> excel data file, for example, deleting the duplicated row, and convert the
> data into csv file. The r program shows the same result. Please kindly advise
> as this problem has bothered for the whole day.
>
> Enclosed is the data set and the respective result. Please kindly assist.
>
> Thanks and regards,
> David Wong
>
> __
> R-help@r-project.org mailing list -- To UNSUBSCRIBE and more, see
> 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.
__
R-help@r-project.org mailing list -- To UNSUBSCRIBE and more, see
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.
Re: [R] High dimensional optimization in R
The postings about polyalgorithms don't mention that optimx has a tool called polyopt() for this. Though I included it in the package, it has not been widely tested or applied, and more experience with such approaches would certainly be of interest to a number of workers, though I suspect the results are rather context-dependent. JN On 2018-12-01 3:52 a.m., Jeremie Juste wrote: > > Hello, > > Genetic algorithm can prove handy as well here. see for instance > https://cran.r-project.org/web/packages/GA/vignettes/GA.html > > with non-convex objective functions I usually try a genetic algorithm for > a few rounds then finish using nlminb > > > Best regards, > Jeremie > > Marc Girondot via R-help writes: > >> I fit also model with many variables (>100) and I get good result when >> I mix several method iteratively, for example: 500 iterations of >> Nelder-Mead followed by 500 iterations of BFGS followed by 500 >> iterations of Nelder-Mead followed by 500 iterations of BFGS >> etc. until it stabilized. It can take several days. >> I use or several rounds of optimx or simply succession of optim. >> >> Marc >> >> Le 28/11/2018 à 09:29, Ruben a écrit : >>> Hi, >>> >>> Sarah Goslee (jn reply to Basic optimization question (I'm a >>> rookie)): "R is quite good at optimization." >>> >>> I wonder what is the experience of the R user community with high >>> dimensional problems, various objective functions and various >>> numerical methods in R. >>> >>> In my experience with my package CatDyn (which depends on optimx), I >>> have fitted nonlinear models with nearly 50 free parameters using >>> normal, lognormal, gamma, Poisson and negative binomial exact >>> loglikelihoods, and adjusted profile normal and adjusted profile >>> lognormal approximate loglikelihoods. >>> >>> Most numerical methods crash, but CG and spg often, and BFGS, >>> bobyqa, newuoa and Nelder-Mead sometimes, do yield good results (all >>> numerical gradients less than 1) after 1 day or more running in a >>> normal 64 bit PC with Ubuntu 16.04 or Windows 7. >>> >>> Ruben >>> >> >> __ >> R-help@r-project.org mailing list -- To UNSUBSCRIBE and more, see >> 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. > > __ > R-help@r-project.org mailing list -- To UNSUBSCRIBE and more, see > 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. > __ R-help@r-project.org mailing list -- To UNSUBSCRIBE and more, see 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.
[R] Question about rmgarch package in R
Good morning, my name is Daniela from Colombia,
I am working with rmgarch package, specifically with the cgarchspec function.
One argument of this function is the "transformation", which is the type of
transformation to apply to the marginal innovations of the GARCH fitted models
(transformation can be parametric, empirical or semi-parametric). Do you know
how this transformations work?. Specially with the parametric transformation,
which distribution is assumed? (normal, t student?), how do I access to this
estimated parameters? This is the code of the function:
cgarchspec(uspec, dccOrder = c(1, 1), asymmetric = FALSE,
distribution.model = list(copula = c("mvnorm", "mvt"),
method = c("Kendall", "ML"), time.varying = FALSE,
transformation = c("parametric", "empirical", "spd"))
Thanks,
Daniela Gualtero
[[alternative HTML version deleted]]
__
R-help@r-project.org mailing list -- To UNSUBSCRIBE and more, see
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.
