Displaying 2 results from an estimated 2 matches for "reshigh".
2013 Sep 27
0
Best and Worst values
...1],use.names=FALSE),Actual=unlist(Actual1[,-1],use.names=FALSE),stringsAsFactors=FALSE)
dat2New<- dat2[!(is.na(dat2$Predict)|is.na(dat2$Actual)),]
?dat3<- dat2New[order(dat2New$S1,dat2New$Predict),]
library(plyr)
resLow<-ddply(dat3,.(S1),summarize, cbind(head(Predict,5),head(Actual,5)))
resHigh<-ddply(dat3,.(S1),summarize, cbind(head(rev(Predict),5),head(rev(Actual),5)))
?resLow1<-data.frame(Date=resLow[,1],Predict=resLow[,2][,1],Actual=resLow[,2][,2])
?resHigh1<-data.frame(Date=resHigh[,1],Predict=resHigh[,2][,1],Actual=resHigh[,2][,2])
?resHigh1$id<- 1:nrow(resHigh1)
?resLow...
2004 Aug 09
4
linear constraint optim with bounds/reparametrization
Hello All,
I would like to optimize a (log-)likelihood function subject to a number of
linear constraints between parameters. These constraints are equality
constraints of the form A%*%theta=c, ie (1,1) %*% 0.8,0.2)^t = 1 meaning
that these parameters should sum to one. Moreover, there are bounds on the
individual parameters, in most cases that I am considering parameters are
bound between zero