About the Left censoring method

[rtrepos]

Hello,
First I would like to thank you for the code you provided on hsmm that works very fine.
I am very interested to apply left censoring on my dataset and as far as I know, your software is the only one to have this option.
You mentioned in your notebook that it is an experimental feature, could you, please, give me more information ? (references, ...)
Thanks in advance
Ronan

[poypoyan]

Hello,

This is just an acknowledgement for now. I will get into this a bit later, because I am not doing machine learning these days.

I will just say two things:

• I am not even sure that "left-censoring" is the correct term for that.
• My reference for both left and right censoring is section 2.2.1. of Yu. The "general assumption" there is when both left and right censoring are ON in my edhsmm.
  
  However, Yu describes a very general hsmm there, so there is a need to simplify the formulas for (explicit duration) hsmm. I need to check again how I implemented left censoring in my code. Then I will add details in my code (esp. _hsmm_core.pyx) and the notebook.

I recommend against using left_censor for serious purposes for now.

Thanks for the feedback!

[rtrepos]

Thank you for your quick answer. It looks promising and deserves to be tested indeed.