Practical Advice collection

mlsa-book / MLSA

https://www.mlsabook.com

MIT License

2 stars 0 forks source link

Practical Advice collection #18

Open adibender opened 11 months ago

adibender commented 11 months ago

don't be more precise than necessary (e.g. when methods use unique events times, data is given very precisely, but time scale of interest is very large, e.g. hdfail example)

RaphaelS1 commented 11 months ago

Ah great issue. Can I also use this to collect FAQs? Like:

At what time scale should predictions be made at? How many unique time intervals to include? (also hdfail). Relevant https://github.com/bcjaeger/aorsf-bench/issues/6#issuecomment-1768137323; https://github.com/imbs-hl/ranger/issues/410#issuecomment-1756745869
Converting predictions between prediction types (crossref composition section)
Something about improper predictions and how this affects predictions and measures

adibender commented 11 months ago

Ah great issue. Can I also use this to collect FAQs? Like:

At what time scale should predictions be made at? How many unique time intervals to include? (also hdfail)

yes, definitely!

adibender commented 8 months ago

For which survival tasks does resampling have to be adjusted, e.g.

recurrent events (by subject or by recurrence)
stratify wrt competing risks
...

adibender commented 7 months ago

Absolut (Probability) vs. relative (hazard) risk, difference and which is needed to answer what questions

RaphaelS1 commented 7 months ago

Absolut (Probability) vs. relative (hazard) risk, difference and which is needed to answer what questions

@adibender when is this relevant?