Calculating loglikelihood of distributions in Python

python statistics distribution log-likelihood
dejoma picture dejoma · May 29, 2018 · Viewed 9.9k times · Source

What is an easy way to calculate the loglikelihood of any distribution fitted to data?

Answer

Cœur picture Cœur · Jul 21, 2018

Solution by OP.

Python has 82 standard distributions which can be found here and in scipy.stats.distributions

Suppose you find the parameters such that the probability density function(pdf) fits the data as follows:

dist = getattr(stats.stats, 'distribution name')
params = dist.fit(data)

Then since it is a standard distribution included in the SciPy library, the pdf and logpdf can be found and used very easily in the following way:

LLH = dist.logpdf(data,*params).sum()

Note that that this corresponds to the loglikelihood function defined here.

Related questions

Fitting empirical distribution to theoretical ones with Scipy (Python)?

INTRODUCTION: I have a list of more than 30,000 integer values ranging from 0 to 47, inclusive, e.g.[0,0,0,0,..,1,1,1,1,...,2,2,2,2,...,47,47,47,...] sampled from some continuous distribution. The values in the list are not necessarily in order, but order doesn't matter for this problem. PROBLEM: Based …

Read More
Two-sample Kolmogorov-Smirnov Test in Python Scipy

I can't figure out how to do a Two-sample KS test in Scipy. After reading the documentation scipy kstest I can see how to test where a distribution is identical to standard normal distribution from scipy.stats import kstest import …

Read More
How to find probability distribution and parameters for real data? (Python 3)

I have a dataset from sklearn and I plotted the distribution of the load_diabetes.target data (i.e. the values of the regression that the load_diabetes.data are used to predict). I used this because it has the …

Read More