We are fitting a lasso model in glmnet as follows:
fit = glmnet(x, y, alpha=1, intercept=F, standardize=F, lower.limits=0)
We were hoping to use selective inference to get p-values on the coefficients. However, there are differences between our model (above) and the given example. For example, we are fitting without an intercept or standardization - but perhaps more importantly, limits on the coefficients. Do you have advice for extending selective inference to the above example? If so, we would really appreciate it. Thank you!
We are fitting a lasso model in glmnet as follows:
fit = glmnet(x, y, alpha=1, intercept=F, standardize=F, lower.limits=0)
We were hoping to use selective inference to get p-values on the coefficients. However, there are differences between our model (above) and the given example. For example, we are fitting without an intercept or standardization - but perhaps more importantly, limits on the coefficients. Do you have advice for extending selective inference to the above example? If so, we would really appreciate it. Thank you!