Coverage increased (+0.8%) to 60.594% when pulling 19436fbfdf408007436a3307ade69cb7050b5129 on aksarkar:fix-pois-log-link into e8a7abc419b58efedca9010ccae5858c8a490876 on stephens999:master.
@aksarkar
i can't remember where we were on this. I remember a question about whether there
was an error in the Lu Thesis. Do you recall this question (and answer to it?)
Add test cases for Poisson-uniform mixture likelihood computations
Compute marginal PMF over a point mass using dpois directly
This fixes issue #114