Hi David,
Thanks for your response. rbind doesnot seem to work.
Here is a reproducible example
Y<-matrix(1:40,ncol=2)
Y1<-Y/60 # estimates of p
#print(Y1)
sigma2<-
matrix(c(var(Y1[,1]),cov(Y1[,1],Y1[,2]),cov(Y1[,1],Y1[,2]),var(Y1[,2])),2,2)
#print(sigma2)
rho<-sigma2[1,2]/sqrt(sigma2[1,1]*sigma2[2,2])
#rho
mean(Y1[,1])
mean(Y1[,2])
#within<-matrix(data=0,nrow=20,ncol=1)
for (rate3 in 1:20){
rate<-Y1[i,]
#print(rate)
rate1<-rate/(1-rate)
rate2<-log(rate1)
Sigma11<-(1/(rate[1]*(1-rate[1]))^2)*sigma2[1,1]
Sigma22<-(1/(rate[2]*(1-rate[2]))^2)*sigma2[2,2]
Sigma12<-(1/((rate[1]*(1-rate[1]))*(rate[2]*(1-rate[2]))))*sigma2[1,2]
Sigma2<-matrix(c(Sigma11,Sigma12,Sigma12,Sigma22),2,2)
#print(Sigma2)
rate3<-mvrnorm(1000, mu=c(rate2[1],rate2[2]), Sigma2)
#print(rate3)
x<-exp(rate3[,1])/(1+exp(rate3[,1]))
y<-exp(rate3[,2])/(1+exp(rate3[,2]))
print(x) # Need to be able to stack the x's to produce one matrix
}
Thanks
Anamika
On Wed, Jul 27, 2016 at 2:46 AM, David Winsemius <dwinsem...@comcast.net>
wrote:
>
> > On Jul 26, 2016, at 8:07 PM, Anamika Chaudhuri <canam...@gmail.com>
> wrote:
> >
> > I have 100 datasets with 20 rows and 2 columns in each dataset.
> > I am looking for help to produce x and y below as 1000 X 20 matrix and
> then
> > repeat that across 100 datasets using R
> >
> > library(MASS)
> > library(car)
> > set.seed(1234)
> > library(mixtools)
> > library(sp)
> >
> > for (k in 1:1){ # k IS THE NO OF DATASETS
> > Y <- read.csv(file=paste0("MVNfreq",k,".csv"))
>
> So this is not reproducible but from the description seems like
>
> do.call( rbind, .... # the list of dataframes might "work" assuming column
> names are _all_ the same.
>
> --
> David.
> >
> > Y<-as.matrix(Y)
> > Y <- ifelse(Y==0,Y+.5,Y)
> >
> >
> > Y1<-Y/60 # estimates of p
> >
> > #print(Y1)
> >
> >
> >
> sigma2<-matrix(c(var(Y1[,1]),cov(Y1[,1],Y1[,2]),cov(Y1[,1],Y1[,2]),var(Y1[,2])),2,2)
> >
> > rho<-sigma2[1,2]/sqrt(sigma2[1,1]*sigma2[2,2])
> > mean(Y1[,1])
> > mean(Y1[,2])
> >
> > #within<-matrix(data=0,nrow=20,ncol=1)
> >
> > for (rate3 in 1:20){
> > rate<-Y1[i,]
> > #print(rate)
> > rate1<-rate/(1-rate)
> > rate2<-log(rate1)
> >
> > Sigma11<-(1/(rate[1]*(1-rate[1]))^2)*sigma2[1,1]
> > Sigma22<-(1/(rate[2]*(1-rate[2]))^2)*sigma2[2,2]
> >
> > Sigma12<-(1/((rate[1]*(1-rate[1]))*(rate[2]*(1-rate[2]))))*sigma2[1,2]
> >
> > Sigma2<-matrix(c(Sigma11,Sigma12,Sigma12,Sigma22),2,2)
> >
> > rate3<-mvrnorm(1000, mu=c(rate2[1],rate2[2]), Sigma2)
> > x<-exp(rate3[,1])/(1+exp(rate3[,1]))
> > y<-exp(rate3[,2])/(1+exp(rate3[,2]))
> > x<-as.data.frame(x)
> > stack(x) # Need help to stack x into a single matrix
> > print(x)
> > print(y)
> > }
> > }
> >
> > [[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.
>
> David Winsemius
> Alameda, CA, USA
>
>
[[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.
