search for: pnlsmaxiter

...ators is at the end of this email. I do have a nagging, unanswered question: What exactly does "Warning message: Step halving factor reduced below minimum in NLS step in: gnls(model = y ~ 5 + ...)" mean? I have tried to address this by specifying "control = list(maxIter = 1000, pnlsMaxIter = 200, msMaxIter = 1000, tolerance = 1e-06, pnlsTol = 1e-04, msTol = 1e-07, minScale = 1e-10, returnObject = TRUE)" in my model calls, but this does not entirely eliminate the problem (I am running gnls( ) 24 separate times on separate data sets). Much thanks in advance, david paul #...

...less than ideal results). The parameters in the function nlmeControl() control the convergence criteria. These default values can be modified to make convergence critera more or less stringent. In an effort to get more converging results, I first increased the iterations for 'maxIter', 'pnlsMaxIter', 'msMaxIter', and 'niterEM' by a factor of 20. I don't believe there is any harm in doing this. There are a few parameters, however, in the nlme controls that I'm not quite sure what they mean or do. For example, exactly what is 'minScale?' The help file says...

...model is parameterized in terms of CL, V1, V2 and Q. basemodel<-nlme(Conc ~TwoCompModel(CL,Q,V1,V2,Time,ID), data = data2, fixed=list(CL+Q+V1+V2~1), random = pdDiag(CL+V1+V2~1), start=c(CL=log(20),Q=log(252),V1=log(24.9),V2=log(120)), control=list(returnObject=TRUE,msVerbose=TRUE, msMaxIter=20,pnlsMaxIter=20,pnlsTol=1), verbose=TRUE). Now, I want to include covariates (age and weight) in the base model. The relationships between CL and covariate is linear and given as the following equation. CL=TVCL+a*wt-b*age, where TVCL is mean clearance, a and b are coefficients of weight and age, respectively....

...= a.max + a.qe + lcp ~ trt, random = a.max + a.qe + lcp ~ 1 | bench/line, start = list(fixed=c(17.4305, 0.182444, 0.00928341, -0.00057221, 44.8384, 8.67678)), method="ML", verbose=T, control = nlmeControl(maxIter=250, msMaxIter=200, pnlsMaxIter=20, gradHess=TRUE, returnObject=TRUE, niterEM=100)) What I'd like to do is to compare reduced models with something like random = list(a.max ~ 1 | bench, a.qe + lcp ~ 1 | bench/line) i.e., to drop the nested term on each of the coefficients one by one so t...

...gle.com/?q=1,A.a2&entry=gmail&source=g>=0,A.a3=0,B=1,B.a2=0,B.a3=0,C=0,C.a2=0,C.a3=0,D=1,D.a2<https://maps.google.com/?q=1,D.a2&entry=gmail&source=g>=0,D.a3=0) try({withTimeout(nlme(model=y~fpl.B.range(x,exp(1),A,B,C,D), control=nlmeControl(maxIter=50,pnlsMaxIter=7,msMaxIter=50,niterEM=25), data=dat, na.action=na.omit, fixed=myFormula,random=list(f.block=pdSymm(A+B+C+D~1)), start=INIT),timeout=10)}) [[alternative HTML version deleted]]

....a3=0) for (i in 1:100) { print(paste("Iteration ",i,"...this will stall soon")) set.seed(i) dat$y <- dat$y+rnorm(nrow(dat), mean = 0, sd = 0.1) try({withTimeout(nlme(model=y~fpl.B.range(x,exp(1),A,B,C,D), control=nlmeControl(maxIter=50,pnlsMaxIter=7,msMaxIter=50,niterEM=25), data=dat, na.action=na.omit, fixed=myFormula,random=list(f.block=pdSymm(A+B+C+D~1)), start=INIT),timeout=10)}) } [[alternative HTML version deleted]]