PanelCount (version 1.0.9)

PoissonRE: A Poisson Model with Random Effects

Description

Estimate a Poisson model with random effects in panel counting data. Note this model is different with the Poisson Lognormal model for counting data.

Usage

PoissonRE(formula, id, data = NULL, par = NULL, sigma = 1,
  max_sigma = 3, method = "BFGS", lower = NULL, upper = NULL, H = 20,
  accu = 10, reltol = 1e-08, verbose = 0, tol_gtHg = Inf)

Value

A list containing the results of the estimated model

Arguments

formula

Formula of the model

id

A vector that represents the identity of individuals, numeric or character

data

Input data, a data frame

par

Starting values for estimates

sigma

Variance of random effects on the individual level

max_sigma

Largest allowed initial sigma

method

Searching algorithm, don't change default unless you know what you are doing

lower

Lower bound for estiamtes

upper

Upper bound for estimates

H

A vector of length 2, specifying the number of points for inner and outer Quadratures

accu

L-BFGS-B only, 1e12 for low accuracy; 1e7 for moderate accuracy; 10.0 for extremely high accuracy. See optim

reltol

Relative convergence tolerance. default typically 1e-8

verbose

Level of output during estimation. Lowest is 0.

tol_gtHg

tolerance on gtHg, not informative for L-BFGS-B

See Also

Other PanelCount: CRE_SS; CRE; PLN_RE; ProbitRE

Examples

# \donttest{
data(rt)
est = PoissonRE(num.words~fans+tweets+as.factor(tweet.id),
                     id=rt$user.id[rt$isRetweet==1],
                     data=rt[rt$isRetweet==1,])
# }