Function to add correlated noise to data

Usage

generateIC(data, corrs, panels, newdata, nsim, dots = FALSE)

Arguments

data: data to be assessed for correlation
corrs: matrix definig the correlation for each panel (each row of matrix) defined in the data
panels: character giving the column of data contaning the information about panel structure
newdata: simulated sets of data (independent)
nsim: number of simulations
dots: (default = FALSE). Logical flag stating whether to show progress dots printed in workspace.

Details

This function is based on work by Iman and Conover (1982).

References

Iman and Conover (1982).

Author

LAS Scott-Hayward, University of St Andrews

Examples

data(nystedA_slim)
nysted$panelid<-as.numeric(nysted$unique.transect.label)

model<-glm(response ~ depth  + as.factor(phase),
              family='poisson', data=nysted)
#> Error in `contrasts<-`(`*tmp*`, value = contr.funs[1 + isOF[nn]]): contrasts can be applied only to factors with 2 or more levels

simData<-generateNoise(n=500, response=fitted(model), family='poisson')
#> Error: object 'model' not found
corrs<-getCorrelationMat(panel = nysted$panelid, data=nysted$response, dots = FALSE)
generateIC(nysted, corrs, "panelid", simData, nsim=10, dots = FALSE)
#> Warning: package 'Hmisc' was built under R version 4.4.3
#> Error: object 'simData' not found