Thread: Calculating Bayes Information Criterion

I know that the formula for the BIC is generally

BIC = -2ln(L) + k*ln(n)

where n is the number of data points, k is the number of parametes, and L = likelihood.

What do we use for L when we're doing BIC for a linear regression or an ANOVA?

Your software should report the likelihood for a linear regression. If it does not, try to run the regression in R or Stata. The estimation command usually reports the likelihood. Try to run in R, It has a special command to compute BIC.

Originally Posted by DannyOcean

I know that the formula for the BIC is generally

BIC = -2ln(L) + k*ln(n)

where n is the number of data points, k is the number of parametes, and L = likelihood.

What do we use for L when we're doing BIC for a linear regression or an ANOVA?