Need help

Dear All,

Thank you very much for all of your help. If anyone feels unhappy with my
words then I apologize.

Actually, I am not an expert user as a result my coding seems to be wrong
on some points. Now, it is working.

I am thankful to all of you for your kind support. Take care.

Hossain



On Fri, Aug 5, 2022 at 4:27 AM Andrew Simmons <akwsimmo at gmail.com> wrote:

Hello again,


Since the GEV distribution and the GPD distribution are similar, I was
able to make dgpd, pgpd, qgpd, and rgpd with no trouble. You should be able
to use the same piece of code I gave earlier to install the package, here
it is again:

install.packages(c("this.path", "essentials"), repos = "
https://raw.githubusercontent.com/ArcadeAntics/PACKAGES")

I tried to make your code a little nicer and also use 'pgpd', let me know
if this behaves as expected.

#####
library(essentials)
a <- rnorm(1000, 110, 5)
b <- rnorm(1000, -0.3, 0.4)
s <- length(a)
lam <- 0.15
thr <- 70
r <- 10

ff <- function(zz) {
     inner <- numeric(s)
     for (k in 1:s) {
         inner[k] <- 1 - lam * pgpd(zz, location = thr, scale = a[k], shape
= b[k], lower.tail = FALSE)
     }


     # if zz is length 1 (it is for uniroot), it would be much faster to do
this:
     # inner <- 1 - lam * pgpd(zz, location = thr, scale = a, shape = b,
lower.tail = FALSE)
     #
     # use whichever you please

     mean(inner) - (1 - 1/r)
}
########
out <- uniroot(ff, lower = 0, upper = 10000)$root
out

On Thu, Aug 4, 2022 at 3:17 PM Md. Moyazzem Hossain <hossainmm at juniv.edu>
wrote:

Dear Andrew Simmons,

Thank you very much.

Actually, I want to calculate the predictive return level of GPD and
compare it with the return level to see the difference. I have used the
following function before but it does not provide the results as expected.
It is OK for GEV but not GPD.

         inner[k] = pgpd(zz, scale = a[k],  shape = b[k] )

Other lines of the code were the same.

Please help me to calculate the predictive return level of GPD.

Thanks in advance.

Hossain

On Thu, Aug 4, 2022 at 5:11 PM Andrew Simmons <akwsimmo at gmail.com> wrote:

Is this a GEV distribution by chance? If so, I made functions dgev,
pgev, qgev, and rgev that you can find install with:

install.packages(c("this.path", "essentials"), repos = "
https://raw.githubusercontent.com/ArcadeAntics/PACKAGES")

and then in your R script

library(essentials)
dgev(...) or pgev or qgev as needed

On Thu, Aug 4, 2022, 09:31 Md. Moyazzem Hossain <hossainmm at juniv.edu>
wrote:

Dear R Experts,

I hope that you are doing well.

I am facing a problem to find out the value of the following function. I
need help in this regard.

#####
a=rnorm(1000, 110, 5)
b = rnorm(1000, -0.3, 0.4)
s = length(a)
lam=0.15
thr=70
r= 10

ff = function(zz){
   inner = vector("numeric", length = s)
      for(k in 1:s){
       inner[k]=(1- lam*((1+b[k]*((zz-thr)/a[k]))^(-1/b[k])))
           }
   answer = mean(inner)- (1- (1/r))
   return(answer)
   }
########
out=uniroot(ff, lower = 0, upper = 10000 )$root
out

########### Error ########
Error in uniroot(ff, lower = 0, upper = 10000) :
   f.upper = f(upper) is NA

Please help me. Thanks in advance.

Take care.

Hossain

--
Best Regards,
Md. Moyazzem Hossain
Associate Professor
Department of Statistics
Jahangirnagar University
Savar, Dhaka-1342, Bangladesh

         [[alternative HTML version deleted]]

--
Best Regards,
Md. Moyazzem Hossain
Associate Professor
Department of Statistics
Jahangirnagar University
Savar, Dhaka-1342, Bangladesh
Website: http://www.juniv.edu/teachers/hossainmm
Research: *[image: Google Scholar]
<https://scholar.google.com/citations?hl=en&user=-U03XCgAAAAJ>* | *ResearchGate
<https://www.researchgate.net/profile/Md_Hossain107>* | *ORCID iD
<https://orcid.org/0000-0003-3593-6936>*