loglik: Log likelihood

In kingaa/pomp: Statistical Inference for Partially Observed Markov Processes

Log likelihood

Description

Extract the estimated log likelihood (or related quantity) from a fitted model.

Usage

logLik(object, ...)

## S4 method for signature 'listie'
logLik(object, ...)

## S4 method for signature 'pfilterd_pomp'
logLik(object)

## S4 method for signature 'wpfilterd_pomp'
logLik(object)

## S4 method for signature 'probed_pomp'
logLik(object)

## S4 method for signature 'kalmand_pomp'
logLik(object)

## S4 method for signature 'pmcmcd_pomp'
logLik(object)

## S4 method for signature 'bsmcd_pomp'
logLik(object)

## S4 method for signature 'objfun'
logLik(object)

## S4 method for signature 'spect_match_objfun'
logLik(object)

## S4 method for signature 'nlf_objfun'
logLik(object, ...)

Arguments

object	fitted model object
...	ignored

Value

numerical value of the log likelihood. Note that some methods compute not the log likelihood itself but instead a related quantity. To keep the code simple, the logLik function is nevertheless used to extract this quantity.

When object is of ‘pfilterd_pomp’ class (i.e., the result of a wpfilter computation), logLik retrieves the estimated log likelihood.

When object is of ‘wpfilterd_pomp’ class (i.e., the result of a wpfilter computation), logLik retrieves the estimated log likelihood.

When object is of ‘probed_pomp’ class (i.e., the result of a probe computation), logLik retrieves the “synthetic likelihood”.

When object is of ‘kalmand_pomp’ class (i.e., the result of an eakf or enkf computation), logLik retrieves the estimated log likelihood.

When object is of ‘pmcmcd_pomp’ class (i.e., the result of a pmcmc computation), logLik retrieves the estimated log likelihood as of the last particle filter operation.

When object is of ‘bsmcd_pomp’ class (i.e., the result of a bsmc2 computation), logLik retrieves the “log evidence”.

When object is of ‘spect_match_objfun’ class (i.e., an objective function constructed by spect_objfun), logLik retrieves minus the spectrum mismatch.

When object is an NLF objective function, i.e., the result of a call to nlf_objfun, logLik retrieves the “quasi log likelihood”.

See Also

Other extraction methods: coef(), cond_logLik(), covmat(), eff_sample_size(), filter_mean(), filter_traj(), forecast(), obs(), pred_mean(), pred_var(), saved_states(), spy(), states(), summary(), time(), timezero(), traces()

kingaa/pomp documentation built on April 19, 2024, 7:12 a.m.