Model-class: An S4 super class to represent a regression-like model

In ktw5691/psychtm: Text Mining Methods for Psychological Research

Description

An S4 super class to represent a regression-like model

Helper function (constructor) for Model class

Usage

## S4 method for signature 'Model'
ndocs(x)

## S4 replacement method for signature 'Model'
ndocs(x) <- value

## S4 method for signature 'Model'
nchain(x)

## S4 replacement method for signature 'Model'
nchain(x) <- value

## S4 method for signature 'Model'
mu0(x)

## S4 replacement method for signature 'Model'
mu0(x) <- value

## S4 method for signature 'Model'
sigma0(x)

## S4 replacement method for signature 'Model'
sigma0(x) <- value

## S4 method for signature 'Model'
eta_start(x)

## S4 replacement method for signature 'Model'
eta_start(x) <- value

## S4 method for signature 'Model'
eta(x)

## S4 replacement method for signature 'Model'
eta(x) <- value

## S4 method for signature 'Model'
loglike(x)

## S4 replacement method for signature 'Model'
loglike(x) <- value

## S4 method for signature 'Model'
logpost(x)

## S4 replacement method for signature 'Model'
logpost(x) <- value

## S4 method for signature 'Model'
waic(x)

## S4 replacement method for signature 'Model'
waic(x) <- value

## S4 method for signature 'Model'
se_waic(x)

## S4 replacement method for signature 'Model'
se_waic(x) <- value

## S4 method for signature 'Model'
p_eff(x)

## S4 replacement method for signature 'Model'
p_eff(x) <- value

## S4 method for signature 'Model'
lpd(x)

## S4 replacement method for signature 'Model'
lpd(x) <- value

## S4 method for signature 'Model'
extra(x)

## S4 replacement method for signature 'Model'
extra(x) <- value

Model(
  ndocs,
  nchain = 1,
  mu0 = NaN,
  sigma0 = NaN,
  eta_start = NaN,
  eta = NaN,
  loglike = NaN,
  logpost = NaN,
  waic = NaN,
  se_waic = NaN,
  p_eff = NaN,
  lpd = NaN
)

Arguments

x	An Model object.
value	A value to assign to a slot for x
ndocs	The number of documents/observations.
nchain	The number of iterations of the Gibbs sampler.
mu0	A (p + 1) x 1 matrix of prior means for eta.
sigma0	A (p + 1) x (p + 1) prior covariance matrix for eta.
eta_start	A (p + 1) x 1 matrix of starting values for eta.
eta	A nchain x (p + 1) matrix of draws of regression coefficients.
loglike	A nchain x 1 vector of the log-likelihood (up to an additive constant).
logpost	A nchain x 1 vector of the log-posterior (up to an additive constant).
waic	WAIC (up to an additive constant) on the deviance scale.
se_waic	Standard error of the WAIC.
p_eff	The effective number of parameters.
lpd	A nchain x ndocs matrix of predictive posterior likelihoods.

Value

A Model object.

Slots

ndocs

The number of documents/observations.

nchain

The number of iterations of the Gibbs sampler.

mu0

A (p + 1) x 1 matrix of prior means for eta.

sigma0

A (p + 1) x (p + 1) prior covariance matrix for eta.

eta_start

A (p + 1) x 1 matrix of starting values for eta.

eta

A nchain x (p + 1) matrix of draws of regression coefficients.

loglike

A nchain x 1 vector of the log-likelihood (up to an additive constant).

logpost

A nchain x 1 vector of the log-posterior (up to an additive constant).

waic

WAIC (up to an additive constant) on the deviance scale.

se_waic

Standard error of the WAIC.

p_eff

The effective number of parameters.

lpd

A nchain x ndocs matrix of predictive posterior likelihoods.

extra

A list of additional model fitting information. Contains time_elapsed, start_time, end_time, corrected_label_switching, and call.

Examples

m1 <- Model(ndocs = 1)
print(m1)

ktw5691/psychtm documentation built on Nov. 3, 2021, 9:10 a.m.