R/defaultExp.R

In SWS-Methodology/faoswsImputation: Package to perform ensemble imputation for various FAO domains

##' The default exponential model for the ensemble model.
##'
##' @param x A numeric vector to be imputed.
##' @export


defaultExp = function(x){

    ### Data Quality Checks
    stopifnot(is.numeric(x))
    stopifnot(length(x) > 1)

    time = 1:length(x)
    yearCount = length(x)
    if(all(is.na(x)))
        return(as.numeric(rep(NA_real_, yearCount)))
    expFit = exp(predict(lm(formula = log(x + 1) ~ time),
                         newdata = data.frame(time = time))) - 1
    ## HACK (Michael): We assign zero to any value that is negative, as it is
    ##                 bounded. This is because the fit gave a value smaller
    ##                 than 1 as the model is uncontraint. A better solution is
    ##                 to have the model bounded.
    expFit[expFit < 0] = 0
    expFit
}