Variance decomposition after running simulations

Dera dynare team

I am estimating a model using stoch_simul for periods 2000, set irf to 100 and Hp filter to 1600 and when it Comes to variances decomposition i get the results but with a note:

Numbers do not add up to 100 due to non zero correlation of sim shochs and II) non linearity.

Is is this something i need to worry about?

Only to some degree. It is simply an explanation that you are using simulation instead of theoretical moments and that your simulation length is finite so that there will be some expected deviations from the theoretically expected asymptotic behavior.

Thanks professor @jpfeifer