find_best_fp_powers {serosv}R Documentation

Returns the powers of the GLM fitted model which has the lowest deviance score.

Description

Refers to section 6.2.

Usage

find_best_fp_powers(data, p, mc, degree, link = "logit")

Arguments

data

the input data frame, must either have 'age', 'pos', 'tot' columns (for aggregated data) OR 'age', 'status' for (linelisting data)

p

a powers sequence.

mc

indicates if the returned model should be monotonic.

degree

the degree of the model. Recommended to be <= 2.

link

the link function. Defaulted to "logit".

Value

list of 3 elements:

p

The best power for fp model.

deviance

Deviance of the best fitted model.

model

The best model fitted

Examples

df <- hav_be_1993_1994
best_p <- find_best_fp_powers(
df,
p=seq(-2,3,0.1), mc=FALSE, degree=2, link="cloglog"
)
best_p


[Package serosv version 1.1.0 Index]