Max LL and Average LL - Githubissues

Labo-Lacourse / stepmix

A Python package following the scikit-learn API for model-based clustering and generalized mixture modeling (latent class/profile analysis) of continuous and categorical data. StepMix handles missing values through Full Information Maximum Likelihood (FIML) and provides multiple stepwise Expectation-Maximization (EM) estimation methods.

https://stepmix.readthedocs.io/en/latest/index.html

MIT License

59 stars 4 forks source link

Max LL and Average LL #55

Closed mohrezazali closed 9 months ago

mohrezazali commented 10 months ago

Hello. Something just grabbed my attention. My model's log_likeligood in each and every iteration was around 70,000 but at the end the average LL was only 3.9676. Can anybody explain how the Average was calculated?

sachaMorin commented 10 months ago

Average LL = LL/(number of observations)

3.96 is your average LL and around 70,000 is your LL. Are you getting different numbers in the progress bar?

sachaMorin commented 9 months ago

@mohrezazali did this answer your question?

sachaMorin commented 9 months ago

Closing. Feel free to reopen if needed!