Paul Zivich

There are two options; multinomial model to predict the distinct categories, or multiple binary models (with some order) Multinomial logit models are looking to be my best bet. Assumes that...

I think I am going to go with the additional pooled logit model, rather than the multinomial. This approach allows for a unique model to be specified for the competing...

Two options with sources for further details here: https://pubmed.ncbi.nlm.nih.gov/19894116-relation-between-three-classes-of-structural-models-for-the-effect-of-a-time-varying-exposure-on-survival/ SNM-AFT numerical solver should be Nelder-Mead SNM-CFT numerical solver should be Newton-Raphson

Ahh I appreciate the heads up! I will take a look at the docs. It would be nice to be able to poll all the current users (or some form...

After looking through `bottleneck`, it doesn't look like it has the tools I need. As mentioned in #78 I might be able to speed up the g-formula by getting a...

C is likely needed to give me the speed boost to get close to SAS's computation time (maybe slightly faster). In case I want to add in the future, some...

The magic-g will not be used by `IterativeCondGFormula` from v0.6.0 onwards. Rather the iterative conditional g-formula takes a list of treatment values. You can also generate a matrix of the...

Unfortunately the implemented AIPW is not ready for survival data. Briefly, AIPW works by combining the IPW with the predicted outcomes values (from g-computation). The formula that combines the two...

`StochasticTMLE` which will be released in v0.8.3 follows the new `verbose` setup. All newly implemented estimators will follow this structure. For the release of v1.0.0, all other estimators will make...

Thanks for looking into this! As for compatibility with DAGitty, I think the format is like below. If that is provided as a string, I think it should be possible...