I’m interested in finding the likelihood of my model as a distribution over parameters. However, pm.traceplot seems to plot the posterior distribution instead of the likelihood distribution. I use a Dirichlet prior, instead of an uniform prior, to constraint the parameters to a simplex. Can I plo…

[image] Hamster_on_wheels: Plot of the log-likelihood ( logp ) as a function of the parameters θ\theta. The plot will not display anything for the values that the sampler hasn’t visited. Not a histogram of log-likelihood values. Not a histogram of the values of the parameters. That is the poste…

Plot the likelihood instead of posterior distribution

Questions

junpenglao May 18, 2019, 2:03pm 5

Bernstein-von Mises theorem is almost never true: https://statmodeling.stat.columbia.edu/2017/11/27/asymptotically-we-are-all-dead/

1 Like

Topic		Replies	Views
How to set up a custom likelihood function for two variables Questions	8	13306	March 9, 2018
Likelihood given by product of two distributions? Questions	6	2215	August 4, 2018
Pymc3.model likelihood evaluation for given prior sample Questions	2	550	June 22, 2018
Computing lppd using trace, Help to understand Questions	4	1212	April 2, 2018
How I get traceplot (or something else) to show priors? Questions	5	5580	June 17, 2019

Plot the likelihood instead of posterior distribution

Related topics