Calculate confidence intervals around parameters of an nlsLoop model fit

confint_nlsLoop.R

# function to get confidence intervals
confint_nlsLoop <- function(x, params, data, id_col, formula, ...){
  dat <- data[data[,id_col] == x,]
  params2 <- params[params[,id_col] == x,]
  fit = NULL
  try(fit <- minpack.lm::nlsLM(formula,
                               data = dat,
                               start = params2[colnames(params2) %in% all.vars(formula[[3]])]), silent = TRUE)
  if(!is.null(fit)){
    confint <- nlstools::confint2(fit)
    confint <- data.frame(confint)
    confint$param <- row.names(confint)
    confint[,id_col] <- x
    colnames(confint)[1:2] <- c('conf.lwr', 'conf.upr')
    rownames(confint) <- NULL
    return(confint)
  }
}

# Example ####

# load in packages
library(nlsLoop) # to install devtools::install_github("padpadpadpad/nlsLoop")
library(nlstools) # to install install.packages("nlstools")

# load in data
data('Chlorella_TRC')

head(Chlorella_TRC)

# run nlsLoop
fits <- nlsLoop(ln.rate ~ schoolfield.high(ln.c, Ea, Eh, Th, temp = K, Tc = 20),
                data = Chlorella_TRC,
                tries = 500,
                id_col = 'curve_id',
                param_bds = c(-30, 30, 0, 10, 0, 30, 245, 385),
                r2 = 'Y',
                supp.errors = 'Y',
                AICc = 'Y',
                na.action = na.omit,
                lower = c(ln.c=-10, Ea=0, Eh=0, Th=0))


# calculate confidence intervals
confidence_intervals <- plyr::ldply(unique(Chlorella_TRC$curve_id),
            confint_nlsLoop,
            fits$params,
            Chlorella_TRC,
            'curve_id',
            fits$formula,
            na.action = na.omit)