search for: objfun

...dCpdt <- -(parms["Vm"]/parms["Vd"])*y[1]/(parms["Km"]+y[1]) list(dCpdt)} Dose<-300 modfun <- function(time,Vm,Km,Vd) { out <- lsoda(Dose/Vd,time,mm.model,parms=c(Vm=Vm,Km=Km,Vd=Vd), rtol=1e-8,atol=1e-8) out[,2] } objfun <- function(par) { out <- modfun(PKindex$time,par[1],par[2],par[3]) sum((PKindex$conc-out)^2) } fit <- optim(c(10,1,80),objfun, method="Nelder-Mead) print(fit$par) [1] 10.0390733 0.1341544 34.9891829 #--Km=0.1341544,wrong value-- #-----wrong model definiens-------- #-----...

Dear list, I'm migrating a project from Matlab to R, and I'm facing a relatively complicated problem for nls. My objective function is below: >> objFun <- function(yEx,xEx,tEx,gamma,theta,kappa){ yTh <- pdfDY(xEx,tEx,gamma,theta,kappa) sum(log(yEx/yTh)^2) } The equation is yTh=P(xEx,tEx) + noise. I collect my data in: >> data <- data.frame(xEx,tEx,yEx) And I do the nls: aux <- nls( ~ objFun(yEx,xEx,tEx,gamma,theta...

...200 Tinf <-0.5   defun<- function(time, y, parms) {  dCpdt <- -parms["kel"] * y[1]  list(dCpdt)  } modfun <- function(time,kel, Vd) {   out <- lsoda(((Dose/Tinf)*(1/(kel*Vd)))*(1-exp(-kel*time)),c(0,time),defun,parms=c(kel=kel,Vd=Vd),rtol=1e-3,atol=1e-5)  out[-1,2]  } objfun <- function(par) {  out <- modfun(PKindex$time, par[1], par[2])  gift <- which( PKindex$conc != 0 )  sum((PKindex$conc[gift]-out[gift])^2)  }        gen<-genoud(objfun,nvars=2,max=FALSE,pop.size=30,max.generations=100,wait.generations=100,starting.value=c(0.7,60),BFGS=FALSE,print.level...

...lution by Emmanuel using least squares. The method (with modification) is as follows - Data1<- data.frame(y=y,x1=x1, x2=x2) # The objective function : least squares. e<-expression((y-(c1+c2*x1+c3*x2))^2) foo<-deriv(e, name=c("c1","c2","c3")) # Objective objfun<-function(coefs, data) { return(sum(eval(foo,env=c(as.list(coefs), as.list(data))))) } # Objective's gradient objgrad<-function(coefs, data) { return(apply(attr(eval(foo,env=c(as.list(coefs), as.list(data))), "gradient"),2,sum)) } D1.unbound<-optim(par=c(c1=0.5, c2=...

Is it possible to use some form of robust regression with the breakpoints routine so that it is less sensitive to outliers? --Rich Richard Kittler Advanced Micro Devices, Inc. Sunnyvale, CA

Hi R-experts, There is lots of information about maximum likelihood estimation in R. However, I didn't came across anything about maximum likelihood with constraints. For example, estimation of parameters k(1) to k(20) with maximum likelihood, where sum(k(i)) = 0. Is there any standard function in R that can do this, or is this something that I should set up myself? Greetings, Church

...myFunction, method = c("L-BFGS-B"), lower = parMin, upper = parMax, control=list(trace=3,factr=iFactr,maxit=300), parOpt=parOpt,parVal=parVal,objFun=iObjFun,dset=dHM1,whatPDM=whatPDMi) Here "myFunction" is a function that provides the value of the sum of squared error that is computed after a lot of computations. There are other parameters and data (through a dataframe dHM1) that are supplied to "myFunction" as given in las...

I have an optimization question that I was hoping to get some suggestions on how best to go about sovling it. I would think there is probably a package that addresses this problem. This is an ordering optimzation problem. Best to describe it with a simple example. Say I have 100 "bins" each with a ball in it numbered from 1 to 100. Each bin can only hold one ball. This optimization is

Greetings, I have a meta-analysis problem in which I have fixed effects regression coefficients (and estimated standard errors) from identical models fit to different data sets. I would like to use these results to create pooled estimated regression coefficients and estimated standard errors for these pooled coefficients. In particular, I would like to estimate the model \beta_{i} = \mu +

...lt;-expression((y-(c1+x1+c2*x2+c3*x3))^2) # Least squares (needs data in env.) > > # Use expression form of deriv(), to allow easy evaluation in a constructed > # environment > > foo<-deriv(e, nam=c("c1","c2","c3")) > > # Objective > > objfun<-function(coefs, data) { +  return(sum(eval(foo,env=c(as.list(coefs), as.list(data))))) + } > > # Objective's gradient > > objgrad<-function(coefs, data) { +  return(apply(attr(eval(foo,env=c(as.list(coefs), as.list(data))), +                        "gradient"),2,s...

...t the docs for method="SANN" (and the examples), you'll see that SANN allows you to pass the "gradient" argument (gr) as a custom function to provide the candidate distribution.  Here's an example: N <- 10 xvec <- seq(0,1,length=N) target <- rank((xvec-0.2)^2) objfun <- function(x) {   sum((x-target)^2)/1e6 } objfun(1:100) swapfun <- function(x,N=10) {   loc <- sample(N,size=2,replace=FALSE)   tmp <- x[loc[1]]   x[loc[1]] <- x[loc[2]]   x[loc[2]] <- tmp   x } set.seed(1001) opt1 <- optim(fn=objfun,               par=1:N,               gr...