osofr / simcausal

Simulating Longitudinal and Network Data with Causal Inference Applications
64 stars 11 forks source link

Bernoulli var affecting Poisson var #21

Open linneafransson opened 7 months ago

linneafransson commented 7 months ago

Hi,

We are currently trying to simulate data for 4 variables. Our issue is that we have 2 variables following Poisson distributions, and we can't find any information about simcausal supporting this distribution.

In our case we have the following DAG with 2 colliders, where we investigate whether source code refactoring influence test code refactoring. SR = Source code refactoring (Bernoulli) TR= Test code refactoring (Bernoulli) R_Loc = Lines of code in child commit (Poisson) R_L = # of locations in child commit (Poisson)

co_tot_effect_v2

Do you have any suggestion regarding how we could achieve the effect of two Bernoulli variables influencing a variable following a Poisson distribution?