statsmodels.distributions.discrete.DiscretizedModel¶
- class statsmodels.distributions.discrete.DiscretizedModel(endog, exog=None, distr=None)[source]¶
experimental model to fit discretized distribution
Count models based on discretized distributions can be used to model data that is under- or over-dispersed relative to Poisson or that has heavier tails.
- Parameters:
- endogarray_like, 1-D
Univariate data for fitting the distribution.
- exog
None
Explanatory variables are not supported. The
exog
argument is only included for consistency in the signature across models.- distr
DiscretizedCount
instance
(required) Instance of a DiscretizedCount distribution.
See also
Examples
>>> from scipy import stats >>> from statsmodels.distributions.discrete import ( DiscretizedCount, DiscretizedModel)
>>> dd = DiscretizedCount(stats.gamma) >>> mod = DiscretizedModel(y, distr=dd) >>> res = mod.fit() >>> probs = res.predict(which="probs", k_max=5)
- Attributes:
endog_names
Names of endogenous variables.
exog_names
Names of exogenous variables.
Methods
expandparams
(params)expand to full parameter array when some parameters are fixed
fit
([start_params, method, maxiter, ...])Fit method for likelihood based models
from_formula
(formula, data[, subset, drop_cols])Create a Model from a formula and dataframe.
get_distr
(params)frozen distribution instance of the discrete distribution.
hessian
(params)Hessian of log-likelihood evaluated at params
hessian_factor
(params[, scale, observed])Weights for calculating Hessian
information
(params)Fisher information matrix of model.
Initialize (possibly re-initialize) a Model instance.
loglike
(params)Log-likelihood of model at params
loglikeobs
(params)Log-likelihood of the model for all observations at params.
nloglike
(params)Negative log-likelihood of model at params
predict
(params[, exog, which, k_max])After a model has been fit predict returns the fitted values.
reduceparams
(params)Reduce parameters
score
(params)Gradient of log-likelihood evaluated at params
score_obs
(params, **kwds)Jacobian/Gradient of log-likelihood evaluated at params for each observation.
Properties
Names of endogenous variables.
Names of exogenous variables.