Solving SDEs in R with Diffeqr package: possible to set seed?

Question

Is it possible to set a seed (like R's set.seed() function) in the diffeqr package in R, while solving stochastic differential equations?

Example

library(diffeqr)
f <- function(u,p,t) {
  return(1.01*u)
}
g <- function(u,p,t) {
  return(0.87*u)
}
u0 = 1/2
tspan <- list(0.0,1.0)
sol = sde.solve(f,g,u0,tspan, alg = "SKenCarp()")
udf = as.data.frame(sol$u)
plotly::plot_ly(udf, x = sol$t, y = sol$u, type = 'scatter', mode = 'lines')

I found in the documentation, that the possibility was present to set a seed, however, if I add it to:

sde.solve(f,g,u0,tspan, alg = "SKenCarp()", seed = 123)

It won't work. Is it possible to set a seed in this R package?

It wasn't part of it before, but https://github.com/JuliaDiffEq/diffeqr/pull/15 adds that feature. If anyone wants to test it we can get it released! — Chris Rackauckas, Sep 19 '19 at 13:38
That should be it, but then you need to go to the PR branch. — Chris Rackauckas, Sep 19 '19 at 13:47
I managed to install it:)! However, somewhere in the `sde.solve()` function it crashes.. I first thought that it might have something to do with the algorithm (SkenCarp), however, also `alg = 'nothing'` (default) crashed with the tutorial example from above. Are there other things I could try? — user213544, Sep 20 '19 at 11:58

score 1 · Accepted Answer · answered Sep 22 '19 at 13:47

library(diffeqr)
diffeqr::diffeq_setup()
f <- function(u,p,t) {
  return(1.01*u)
}
g <- function(u,p,t) {
  return(0.87*u)
}
u0 = 1/2
tspan <- list(0.0,1.0)
sol = sde.solve(f,g,u0,tspan, alg = "SKenCarp()", seed=1)
udf = as.data.frame(sol$u)
plotly::plot_ly(udf, x = sol$t, y = sol$u, type = 'scatter', mode = 'lines')

Now works on diffeqr v0.1.3, which was just submitted to CRAN.

BTW, SKenCarp probably isn't the right method for this problem, but I assume you're testing it for other reasons.

Solving SDEs in R with Diffeqr package: possible to set seed?

1 Answers1