Call function in redefining the function in R - r

Question

I want to redefine my function in a loop by calling the function from last iteration. However I know this is basically a recursive way which I don't want. To give an example, see the following
for(i in 1:3)
{
...
myfunction<-function(y){myfunction(y)*dnorm(full_x[j],mean=y*full_x[j-1],sd=true_eps)}
...
result[[i]]<-myfunction
}
What I want is that in step t, I define a function called myfunction and in step t+1, I want to redefine the function by using myfunction from the last step in order to do some dynamic updatings. The program keeps giving me the error: "Error: C stack usage 7971152 is too close to the limit". I guess the problem lies in this "recursive way". I tried to relabel the function in each loop but it did not work. Is there any ways to solve this?
**A concrete example would be the following
# loop for calculating the density function
for(j in 2:length(full_x))
{
# define the normal density
trued<-function(y){dnorm(full_x[j],mean=y*full_x[j-1],sd=true_eps)}
# joint distribution
# which is normal density multiplied by a piror
if(j>=3)
{
trued<- function(y){trued(y)**true_density(y)}
}
# integration of the density w.r.t. rho
trueint<-integrate(trued, lower = 0, upper = 1)$value
# density function
true_density<-function(y) {trued(y)/trueint}
# save into list
dyn_density_true[[j-1]]<-true_density
}
What I want to do is to calculate a sequence of density functions. In the first step, it is just a normal density weighed by its integral (so posterior density). In the following steps, I need to use the density function from last step as a prior then do the same exercise again.

Answer 1

You're referring to myfunction directly when you want to refer to the latest entry in result. If you print myfunction or result you'll see why it doesn't work.
You can instead create a string with the actual function you want and then parse it. I've created a sample function below where each iteration multiplies the previous function result by 5.
f <- function(y) y**2
k <- 5
res <- list(f)
for (i in 1:3) {
fstring <- paste0('function(y) k*res[[',i,']](y)')
res[[i+1]] <- eval(parse(text = fstring))
}
res
#> [[1]]
#> function(y) y**2
#>
#> [[2]]
#> function(y) k*res[[1]](y)
#>
#> [[3]]
#> function(y) k*res[[2]](y)
#>
#> [[4]]
#> function(y) k*res[[3]](y)
lapply(res,function(f) f(2))
#> [[1]]
#> [1] 4
#>
#> [[2]]
#> [1] 20
#>
#> [[3]]
#> [1] 100
#>
#> [[4]]
#> [1] 500
Created on 2019-10-16 by the reprex package (v0.3.0)