Learn R Programming

pomp (version 6.1)

as.data.frame: Coerce to data frame

Description

All pomp model objects can be recast as data frames. The contents of the resulting data frame depend on the nature of the object.

Usage

# S3 method for pomp
as.data.frame(x, ...)

# S3 method for pfilterd_pomp as.data.frame(x, ...)

# S3 method for probed_pomp as.data.frame(x, ...)

# S3 method for kalmand_pomp as.data.frame(x, ...)

# S3 method for bsmcd_pomp as.data.frame(x, ...)

# S3 method for pompList as.data.frame(x, ...)

# S3 method for pfilterList as.data.frame(x, ...)

# S3 method for abcList as.data.frame(x, ...)

# S3 method for mif2List as.data.frame(x, ...)

# S3 method for pmcmcList as.data.frame(x, ...)

# S3 method for wpfilterd_pomp as.data.frame(x, ...)

Arguments

x

any R object.

...

additional arguments to be passed to or from methods.

Details

When object is a simple ‘pomp’ object, as(object,"data.frame") or as.data.frame(object) results in a data frame with the times, observables, states (if known), and interpolated covariates (if any).

When object is a ‘pfilterd_pomp’ object, coercion to a data frame results in a data frame with the same content as for a simple ‘pomp’, but with conditional log likelihood and effective sample size estimates included, as well as filtering means, prediction means, and prediction variances, if these have been computed.

When object is a ‘probed_pomp’ object, coercion to a data frame results in a data frame with the values of the probes computed on the data and on simulations.

When object is a ‘kalmand_pomp’ object, coercion to a data frame results in a data frame with prediction means, filter means and forecasts, in addition to the data.

When object is a ‘bsmcd_pomp’ object, coercion to a data frame results in a data frame with samples from the prior and posterior distribution. The .id variable distinguishes them.

When object is a ‘wpfilterd_pomp’ object, coercion to a data frame results in a data frame with the same content as for a simple ‘pomp’, but with conditional log likelihood and effective sample size estimates included.