3.7.2. statsmodels.miscmodels.count¶
Created on Mon Jul 26 08:34:59 2010
Author: josef-pktd
changes: added offset and zero-inflated version of Poisson
- kind of ok, need better test cases,
- a nan in ZIP bse, need to check hessian calculations
- found error in ZIP loglike
- all tests pass with
3.7.2.1. Issues¶
If true model is not zero-inflated then numerical Hessian for ZIP has zeros for the inflation probability and is not invertible. -> hessian inverts and bse look ok if row and column are dropped, pinv also works
- GenericMLE: still get somewhere (where?)
“CacheWriteWarning: The attribute ‘bse’ cannot be overwritten”
bfgs is too fragile, doesn’t come back
nm is slow but seems to work
need good start_params and their use in genericmle needs to be checked for consistency, set as attribute or method (called as attribute)
numerical hessian needs better scaling
check taking parts out of the loop, e.g. factorial(endog) could be precalculated
3.7.2.3. Classes¶
GenericLikelihoodModel(endog[, exog, ...]) |
Allows the fitting of any likelihood function via maximum likelihood. |
NonlinearDeltaCov(fun, params, cov_params) |
Asymptotic covariance by Deltamethod |
PoissonGMLE(endog[, exog, loglike, score, ...]) |
Maximum Likelihood Estimation of Poisson Model |
PoissonOffsetGMLE(endog[, exog, offset, missing]) |
Maximum Likelihood Estimation of Poisson Model |
PoissonZiGMLE(endog[, exog, offset, missing]) |
Maximum Likelihood Estimation of Poisson Model |