scaledlogit: Scale Parameter for Integer Modeling and Forecast

In JobNmadu/Dyn4cast: Dynamic Modeling and Machine Learning Environment

Scale Parameter for Integer Modeling and Forecast

Description

This function is a wrapper for scaling the fitted (predicted) values of a one-sided (positive or negative only) integer response variable of supported models. The scaling involves some log transformation of the fitted (predicted) values.

Usage

scaledlogit(x, lower, upper)

Arguments

x	The parameter to be scaled, which is the fitted values from supported models. The scaled parameter is used mainly for constrained forecasting of a response variable positive (0 - inf) or negative (-inf - 0). The scaling involves log transformation of the parameter
lower	Integer or variable representing the lower limit for the scaling (-inf or 0)
upper	Integer or variable representing the upper limit for the scaling (0 or inf)

Examples

library(Dyn4cast)
library(splines)
lower <- 1
upper <- 37
Model   <- lm(states ~ bs(sequence, knots = c(30, 115)), data = Data)
scaledlogit(x = fitted.values(Model), lower = lower,
 upper = upper)

JobNmadu/Dyn4cast documentation built on March 5, 2025, 9:56 p.m.