kgoldfeld

I totally missed that you were with the IBM/python team. The question makes much more sense. I am afraid you are going to find a lot of these weird things...

Yes - that is the intended outcome. You might want to do that if you follow up by generating a time-dependent outcome. Here is a simple example: ``` defT 2:...

I definitely wasn't thinking of implementing exactly like this - and the use of `AUC::auc()` was merely to check the results - it is not part of the algorithm. There...

I am not saying this is the most crucial thing, but what if we just added two new distributions that happen to act just like two existing distributions? So, we...

But what if I said we want to introduce a new distribution, say the Weibull distribution - would you make the same argument?

This is interesting. It would represent a different fundamental understanding of the of formula, as the argument income is the entire vector (rather than a row-specific value). I know we...

I agree - it is nice to have a motivating reason to do this, other than the fact that using "variance" to refer to a dispersion parameter or, even worse,...

The first call to a defData (when creating a new definition) is creating the first variable, there is no way to reference an existing variable in the data.table. So, the...

This is in line with what I had been thinking. In concept, I like the 2nd option you just mentioned. But, as I think about it, one issue I see...

Not dumb - it was just a highly simplified example - so my fault. Typically, it would look more like this, where there is some outcome variable that is a...