Constrained Optimization in Rstan

[braggbear]

Hello. I am trying to run a constrained optimization in rstan. My model looks something like y ~ B0 + B1X1 + B2X2. I already built a model to find the posterior estimates for B0, B1, and B2. Now I want to find the optimal values of X1 and X2 in order to maximize y (given some constraints on how big or small the X’s can be).

Can someone point me to some example code on how to do this? Much appreciated.

Thanks!

[torkar]

Hi,

could you perhaps use posterior_linpred() to see at which predictor values y has its maximum value? If you want to propagate the uncertainty perhaps posterior_predict() could be used instead?

Reproduce posterior_predict from posterior_linpred for a neg binom spline model

Hi Angelo, here’s a brief explanation of the difference. Suppose you have a negative binomial regression with a log link: \eta_i = a + bx_i \\ \lambda_i = \exp(\eta_i) \\ y_i \sim {\rm NegBinom}(\lambda_i, \phi) posterior_linpred() gives you posterior draws of \eta. posterior_linpred(transform=TRUE), or equivalently posterior_epred(), gives you posterior draws of \lambda, that is, it does the inverse link transformation for you. posterior_predict() gives you draws from {\rm NegBinom}(\l…