R/process_length_data.R

In ChristineStawitz-NOAA/stspsize: State-Space Autoregressive Length-at-Age Model

#' Process length_data A function that takes processed length data and
#' standardizes it to calculate the means.
#'
#' @param spp_data a tibble that has, at minimum, the columns `age_years`,
#'   `length_cm`, `year`
#' @param common_ Common name of the species you are interested in
#' @param minimum_n The minimum sample size used to set the maximum age of fish
#'   observed
#' @param plot_bool TRUE or False should plots be made
#' @param ... Other arguments to pass to [get_std_length()]
#'
#' @return the processed length data
#' @export
process_length_data <- function(spp_data, common_, minimum_n,
                                plot_bool = FALSE, ...) {

  # Get mean stats
  mean_mat <- get_std_length(spp_data, ...)

  #Get sample variance
  summ <- 0
  n <- nrow(spp_data)
  for(i in nrow(spp_data)){
    summ <- summ + (spp_data$length_cm[i] - mean_mat$meanl[which(mean_mat$age_years==spp_data$age_years[i])])^2
  }
  sample_var <- summ/(n-1)

  # Get rid of ages with fewer than minimum_n obs
  ages <- filter(mean_mat, count < minimum_n) %>% select(age_years)

  # process the data
  # 1. join data with mean matrix,
  # 2. remove ages with small sample size
  # 3. group by ages and years 4. get mean length for each age-year combo
  # 5. get difference between age-year combo and overall mean length for ages
  processed_data <- left_join(spp_data, mean_mat, by = "age_years") %>%
    filter(!(age_years %in% ages)) %>%
    filter(sdl > 0)

  if (plot_bool) {
    plot_data <- processed_data %>%
      mutate(standardl = (length_cm - meanl) / sdl)
    length_plots(plot_data, name = common_)
  }
  processed_data <- processed_data %>%
    group_by(age_years, year) %>%
    mutate(meanal = mean(length_cm)) %>%
    mutate(standardl = (meanal - meanl) / sdl)
  processed_data
}