smooth.OMG

class smooth.OMG(model_a='MNN', model_b=None, lags=None, orders_a=None, orders_b=None, constant_a=False, constant_b=None, formula_a=None, formula_b=None, regressors_a='use', regressors_b=None, persistence_a=None, persistence_b=None, phi_a=None, phi_b=None, arma_a=None, arma_b=None, h=0, holdout=False, initial='backcasting', loss='likelihood', reg_lambda=None, ic='AICc', bounds='usual', verbose=0, nlopt_kwargs=None, ets='conventional', smoother='default')

General occurrence model — two parallel ETS sub-models combined.

Public API:: fit(y, X=None) → self predict(h, X=None) → ForecastResult
Attributes after fit:: model_a : OM — odds-ratio sub-model model_b : OM — inverse-odds-ratio sub-model fitted : combined probabilities ∈ (0, 1) residuals : ot - fitted loss_value, loglik, aic/aicc/bic/bicc coef : joint parameter vector concat(B_A, B_B) model_name : "oETS[G](MNN)(MNN)"-style string

Methods

`coefbootstrap`([nsim, size, replace, prob, ...])	Bootstrap the joint OMG coefficient sampling distribution.
`confint`([parm, level, type, bootstrap, ...])	Confidence intervals for the joint OMG parameters.
`fit`(y[, X])
`multicov`([type, h, nsim])	Multi-step forecast-error covariance — not defined for OMG.
`point_lik`([log])	Per-observation log-likelihood of the coupled occurrence model.
`predict`(h[, X, interval, level, side])
`rstandard`()	Pearson standardised residuals for the general occurrence model.
`rstudent`()	Pearson studentised residuals for the general occurrence model.
`simulate`([nsim, seed, obs])	Re-simulate the combined OMG occurrence series.
`summary`([level, digits, type])	Coefficient-table summary for the joint occurrence model.
`vcov`([type, bootstrap, step_size])	Joint variance–covariance matrix for both OMG sub-models.

Attributes

`actuals`	Binary 0/1 indicator built from the raw input series.
`aic`
`aicc`
`b_value`
`bic`
`bicc`
`coef`
`coef_names`	Joint parameter names with `A:` / `B:` prefixes.
`distribution_`
`fisher_information_`	Cached observed Fisher Information at the estimated joint B.
`fitted`
`holdout_data`
`lags_used`
`loglik`
`loss_`
`loss_value`
`model`
`model_name`
`nobs`
`nparam`
`occurrence`
`residuals`
`scale`
`sigma`
`time_elapsed`