[ANN] Update to SolveDSGE

Package Announcements
7

I think some of this is due to the default: burn_in = 1000 currently hard-coded in all stochastic simulations.

1 It could be nice to keep burn_in = 1000 as the default but also to allow the user to customize.
For example:

g(x; burn_in=1000) = x + burn_in
g(5) #1005
g(5, burn_in=3) #8

Here:

function simulate(soln::R,initial_state::Array{T,1},n::S; burn_in=1000)

@sglyon does something similar: (CLMMJuliaPythonMatlab/sim_resids.jl at master · sglyon/CLMMJuliaPythonMatlab · GitHub) setting default nburn=200

2 Similarly it could be nice to allow user to customize the seed or possibly generate multiple simulations

function simulate(soln::R,initial_state::Array{T,1},n::S; burn_in=1000, seed=123456)

3 currently the only option for generating shocks is randn this can also be customized or maybe even allow a different distribution for each shock?

I submitted a PR for burn_in. This won’t cause any breaks bc the default burn_in remains the same. Likewise the default arguments for simulate remain: simulate(soln,initial_state,n)