# Averaging multiple runs using Gillepsie.jl

**URL:** https://discourse.julialang.org/t/averaging-multiple-runs-using-gillepsie-jl/40359
**Category:** New to Julia
**Created:** [May 28, 2020, 5:00pm UTC](https://discourse.julialang.org/t/averaging-multiple-runs-using-gillepsie-jl/40359 "2020-05-28T17:00:11Z")
**Posts on this page:** 1
**Page:** 1

<div class="post-metadata">

### Author: ![erlebach](https://sea2.discourse-cdn.com/julialang/user_avatar/discourse.julialang.org/erlebach/32/12973_2.png) [@erlebach](https://discourse.julialang.org/u/erlebach)
#### Post date: [May 28, 2020, 5:00pm UTC](https://discourse.julialang.org/t/averaging-multiple-runs-using-gillepsie-jl/40359/1 "2020-05-28T17:00:12Z")

</div>

I am continuing with my experiments with the SIR model. I would like to average the results from multiple runs using the `ssa` function in `Gillepsie.jl`. This is not straightforward since jumps are used, which are stochastic. Is tehre an easy way to accomplish this?

One way would be to use a package that bins the results into a fixed number of bins, which would permit averaging the multiple solutions. I know of various binning options if one has a single array. I want to bin all the columns of a database according to the first column, which are the time stamps (a float). Any suggestions?

Next, I will try to use DifferentialEquations.jl directly using the Jump package (which is not as elegant as Gillepsie in terms of problem set up, IMHO). I plan to run more complex models at a later time. I assume that the JumpProblem solvers have time interpolation built in so that I can interpolate the solution onto a fixed time discretization prior to averaging?