search for: maxbfgs

Dear all, I need to maximize the v: v= D' W D D is a column vector ( n , 1) W is a given matrix (n, n) subject to: sum D= 1 (BTW, n is less than 300) I´ve tried to use maxBFGS, as follows: ##################################### objectiveFunction<-function(x) { return(t(D)%*%W%*%D) } Amat<-diag(nrow(D)) Amat<-rbind((rep(-1, nrow(D))), Amat) bvec<-matrix( c(0), nrow(D)+1, 1) bvec[1,1]<-c(1) startValues=rep(1/nrow(D),nrow(D)) #Istart value is homogeneous d...

...],-1)^((-2)*(alpha+1)) )}); s2<- sum(for(m in 1:n){log(lag(c[m],-1)^(((2)*alpha)+2))}); constant-(74*s1)-(1/2)*s2} In particular I want to use the BHHH and the BFGS algorithms. I have found the following packages that can be used for the task: *mle{stats4}, optim{stats}, maxBHHH{micEcon}, maxBFGS{micEcon}*. I have documented some of my attempts below ((a) package name (b) usage (c) my attempt and corresponding error). In all humility I apologise for any bad coding, and ask if anyone can *direct me in finding these estimators*. Yours sincerely. *(1a) mle{stats4} (b) Usage: m...

...;>>>> Steven Yen >>>>> on Fri, 9 Oct 2020 05:39:48 +0800 writes: > Oh Hi Arne, You may recall we visited with this before. I > do not believe the problem is algorithm specific. The > algorithms I use the most often are BFGS and BHHH (or > maxBFGS and maxBHHH). For simple econometric models such > as probit, Tobit, and evening sample selection models, old > and new versions of R work equally well (I write my own > programs and do not use ones from AER or > sampleSekection). For more complicated models the newer R...

...maximum likelihood functions a short while ago and got an email saying that some of the content was "unprocessed". Hence, I am reposting the question just to be sure (sorry for the multiple emails if both reached you). My question is reagarding the way we use constraints in programs like maxBFGS, etc. We need to put two matrices into a list, containing the mutipliers on the LHS of the constraint equations and the constants on the RHS. I want to constrain the last parameter of my routine to a negative value. Hence, I specify: lhs=matrix(0,nrow=noOfParameters,ncol=noOfParameters) lhs[noOfP...

Hi all, Sorry to bother you. I'm estimating a discrete choice model in R using the maxBFGS command. Since I wrote the log-likelihood myself, in order to double check, I run the same model in Limdep. It turns out that the coefficient estimates are quite close; however, the standard errors are very different. I also computed the hessian and outer product of the gradients in R using the num...

Oh Hi Arne, You may recall we visited with this before. I do not believe the problem is algorithm specific. The algorithms I use the most often are BFGS and BHHH (or maxBFGS and maxBHHH). For simple econometric models such as probit, Tobit, and evening sample selection models, old and new versions of R work equally well (I write my own programs and do not use ones from AER or sampleSekection). For more complicated models the newer R would converge with not-so-nice gra...

...ind(Pn,Pmn) } Pm<-rowMeans(Pn) /* The simulated probabilities for one person P<-rbind(P,Pm) /* Obtain the choice probabilities for all 1,000 people } sum(log(P)*(as.numeric(y))) /* Log-likelihood function, where y is the variable for choices } A<-matrix(c(0,1,0),1,3) B<-0 rp<-maxBFGS(RP,start=c(-0.072,0.8,0.539),y=Y,X=EV,constraints=list(ineqA=A,ineqB=B)) /* Constrained MLE, to ensure that the estimated standard deviation is positive; the start values are taken from conditional logit estimation -- Min Chen Graduate Student Department of Agricultural, Food, and Resource Eco...

Hi Steven Which optimisation algorithms in maxLik work better under R-3.0.3 than under the current version of R? /Arne On Thu, 8 Oct 2020 at 21:05, Steven Yen <styen at ntu.edu.tw> wrote: > > Hmm. You raised an interesting point. Actually I am not having problems with aod per se?-it is just a supporting package I need while using old R. The essential package I need, maxLik, simply

..., A2, EF, DF, df1, df2) rowi3 = c(Fold, Fnew, f, Fold < f, Fnew < f) row[1,] = c(rowi1,rowi2,rowi3) write.table(row,file="70addadd__goods.csv",append=T,quote=T,sep=";",row.names=F,col.names=F) rm(y, NR, kr_nr, NRcode, kr_nrF_old, kr_nrF_new, qff) } else { BFGS <- try(maxBFGS(loglk(simtype),gradvec(simtype),hessmat(simtype),start=starter(simtype,x,y)),silent=TRUE) if (is.list(BFGS) == TRUE) kr_bfgs <- kenward_roger_both(simtype,BFGS$estimate,beta_tested,x,y,z) if (is.list(BFGS) == TRUE) BFGScode <- BFGS$code else BFGScode <- 99 if (existencecheck(kr_bfgs) ==...