R/npv_irr.R

In forestmangr: Forest Mensuration and Management

#' @title 
#' Calculate Net Present Value and other economic variables
#' @description 
#' Get the net present value, internal rate of return, and other 
#' economic variables, given cost and revenue values.
#' 
#' @param df A data frame.
#' @param year Quoted name of the year variable.
#' @param cost Quoted name of the costs variable.
#' @param revenue Quoted name of the revenue variable.
#' @param rate Numeric value of the yearly rate to be used.
#' @param output Selects different output options. It can be either \code{"full"} for a list containing a sensibility plot and a data frame with a single observation and one column for each variable, or \code{"simple"} for a two column data frame with one observation for each calculated variable. Default: \code{"simple"}.
#' @param sens_limits Selects the rate rage used in the sensibility plot. This is a numeric vector with two elements, the initial and final rate to be used as range. These can vary between 0 and 100. Default: \code{c(1,30)}.
#' @param big_mark Selects thousands separator. Can be either \code{"."}, \code{" "} or \code{","}. Default: \code{","}.
#' @param dec_mark Selects decimal separator. Can be either \code{","} or \code{"."}. Default: \code{"."}.
#' @param prefix selects the prefix for the y axis in the sensibility plot. Can be either \code{"$"} or \code{"R$"}. Default: \code{"$"}.
#' @return A data frame, or a list, according to output.
#' @export
#' @examples 
#' \dontrun{
#' library(forestmangr)
#' data(exfm22)
#' 
#' npv_irr(exfm22,"year","cost","revenue",rate=8.75)
#'
#' # To also get a sensibility plot, use
#' 
#' npv_irr(exfm22,"year","cost","revenue",rate=8.75, output="full")
#' } 
#' @author Sollano Rabelo Braga \email{sollanorb@@gmail.com}

npv_irr <- function(df, year, cost, revenue, rate, output="full", sens_limits = c(1,30), big_mark=",", dec_mark=".", prefix="$"){
  # ####
  . <- VoCT_total <- VoRT_total <- VnCT_total <- VnRT_total <- n <- IRR <- Value <- Variable <- x_axis <- VPR <- VPC <- VFR <- VFC <- VPL <- VET <- NPV <- ELV <- facet_var <- y_axis <- red_y2_axis <- red_y_axis <- NULL
  # ####
  
  # se df nao for fornecido, nulo, ou  nao for dataframe, ou nao tiver tamanho e nrow maior que 1,parar
  if(  missing(df) ){  
    stop("df not set", call. = F) 
  }else if(!is.data.frame(df)){
    stop("df must be a dataframe", call.=F)
  }else if(length(df)<=1 | nrow(df)<=1){
    stop("Length and number of rows of 'df' must be greater than 1", call.=F)
  }
  
  # se year nao for fornecido nao for character, ou nao for um nome de variavel,ou nao for de tamanho 1, parar
  if(  missing(year) ){  
    stop("year not set", call. = F) 
  }else if( !is.character(year) ){
    stop("'year' must be a character containing a variable name", call.=F)
  }else if(length(year)!=1){
    stop("Length of 'year' must be 1", call.=F)
  }else if(forestmangr::check_names(df, year)==F){
    stop(forestmangr::check_names(df, year, boolean=F), call.=F)
  }
  
  # se cost nao for fornecido nao for character, ou nao for um nome de variavel,ou nao for de tamanho 1, parar
  if(  missing(cost) ){  
    stop("cost not set", call. = F) 
  }else if( !is.character(cost) ){
    stop("'cost' must be a character containing a variable name", call.=F)
  }else if(length(cost)!=1){
    stop("Length of 'cost' must be 1", call.=F)
  }else if(forestmangr::check_names(df, cost)==F){
    stop(forestmangr::check_names(df, cost, boolean=F), call.=F)
  }
  
  # se revenue nao for fornecido nao for character, ou nao for um nome de variavel,ou nao for de tamanho 1, parar
  if(  missing(revenue) ){  
    stop("revenue not set", call. = F) 
  }else if( !is.character(revenue) ){
    stop("'revenue' must be a character containing a variable name", call.=F)
  }else if(length(revenue)!=1){
    stop("Length of 'revenue' must be 1", call.=F)
  }else if(forestmangr::check_names(df, revenue)==F){
    stop(forestmangr::check_names(df, revenue, boolean=F), call.=F)
  }
  
  # Se rate nao for numerico, nao for de tamanho 1, ou nao estiver dentro dos limites, parar
  if(!is.numeric( rate )){
    stop( "'rate' must be numeric", call.=F)
  }else if(length(rate)!=1){
    stop("length of 'rate' must be 1", call.=F)
  }else if(! rate > 0 | ! rate <= 1000){
    stop("'rate' must be a number between 0 and 1000", call.=F)
  }
  
  # Se output nao for character,ou nao for de tamanho 1, parar
  if(!is.character( output )){
    stop( "'output' must be character", call.=F)
  }else if(length(output)!=1){
    stop("Length of 'output' must be 1", call.=F)
  }else if(! output %in% c('full', 'simple') ){ 
    stop("'output' must be equal to 'full' or 'simple' ", call. = F) 
  }
  
  # Se sens_limits nao for numerico, nao for de tamanho 1, ou nao estiver dentro dos limites, parar
  if(!is.numeric( sens_limits )){
    stop( "'sens_limits' must be numeric", call.=F)
  }else if(length(sens_limits)!=2){
    stop("length of 'sens_limits' must be 2", call.=F)
  }else if(! sens_limits[1] >= 0 | ! sens_limits[1] <= 100){
    stop("'sens_limits' first element must be a number between 0 and 100", call.=F)
  }else if(! sens_limits[2] >= 0 | ! sens_limits[2] <= 100){
    stop("'sens_limits' second element must be a number between 0 and 100", call.=F)
  }else if(sens_limits[1] > sens_limits[2]){
    stop("'sens_limits' first element cannot be greater than the second", call.=F)
  }
  
  # Se big_mark nao for character,ou nao for de tamanho 1, parar
  if(!is.character( big_mark )){
    stop( "'big_mark' must be character", call.=F)
  }else if(length(big_mark)!=1){
    stop("Length of 'big_mark' must be 1", call.=F)
  }else if(! big_mark %in% c('.', ',', " ") ){ 
    stop("'big_mark' must be equal to '.', ',' or ' ' ", call. = F) 
  }
  
  # Se dec_mark nao for character,ou nao for de tamanho 1, parar
  if(!is.character( dec_mark )){
    stop( "'dec_mark' must be character", call.=F)
  }else if(length(dec_mark)!=1){
    stop("Length of 'dec_mark' must be 1", call.=F)
  }else if(! dec_mark %in% c('.', ',') ){ 
    stop("'dec_mark' must be equal to '.' or ',' ", call. = F) 
  }
  
  # ####
  year_name <- year
  year_sym <- rlang::sym(year)
  cost_sym <- rlang::sym(cost)
  revenue_sym <- rlang::sym(revenue)
  
  df[is.na(df)] <- 0
  
  tab1 <- df %>% 
    dplyr::mutate(
      !!year_name := as.integer(!!year_sym), # converter pra integer, pra diferenciar dos outros numeros, que sao  double
      saldo = (!!revenue_sym) - (!!cost_sym),
      VoCT = (!!cost_sym)/(1+rate/100)^(!!year_sym),
      VoRT = (!!revenue_sym)/(1+rate/100)^(!!year_sym), 
      VnCT = (!!cost_sym)*(1+rate/100)^(max(!!year_sym,na.rm=TRUE) - (!!year_sym) ),
      VnRT = (!!revenue_sym)*(1+rate/100)^(max(!!year_sym,na.rm=TRUE) - (!!year_sym) )) %>% 
    dplyr::mutate_if(is.double,formattable::comma,big.mark=big_mark,decimal.mark=dec_mark) # formatar todos os numeros que sao double
  
  tab1
  
  # Calcula tir
  irr <- FinCal::irr2(as.numeric(tab1$saldo))
  
  
  tab2 <- tab1 %>%
    dplyr::summarise_all(list(total=sum(.,na.rm=TRUE))) %>% 
    dplyr::mutate(n=max(df[[year_name]]),
                  #rate = formattable::percent(rate/100),
                  !!paste(year_name,"total",sep="_"):=NULL,
                  NPV  = VoRT_total - VoCT_total, # Net Present Value
                  NFV  = VnRT_total - VnCT_total, # Net Future Value
                  CB   = VoRT_total / VoCT_total, # Cost benefit Ratio
                  ELV  = (VnRT_total - VnCT_total) / ((1+rate/100)^n - 1 ), # Expected Land Value
                  YNPV = ((VnRT_total - VnCT_total)*rate/100) / ((1+rate/100)^n - 1 ), # Yearly Net Present Value
                  IRR  = formattable::percent(irr,decimal.mark=dec_mark)  ) #Internal Rate of Return
  
  if(output=="simple"){
    
    return(dplyr::select(tab2, -dplyr::contains("total"), -n) )
    
  }else if(output=="full"){
    suppressWarnings(
      
      tab2_mod <- tab2 %>% 
        dplyr::mutate_at(
          dplyr::vars(
            #rate,
            IRR),~.*100) %>% 
        dplyr::select(-dplyr::contains("total"),
                      #-rate,
                      -n) %>% 
        tidyr::gather("Variable", "Value") %>% 
        dplyr::mutate_at(
          dplyr::vars(Value),
          formattable::comma,big.mark=big_mark,decimal.mark=dec_mark )
    )
    
    tab2_mod$Variable <- c("Net Present Value (NPV)",
                           "Net Future Value (NFV)",
                           "Cost benefit Ratio (CB)",
                           "Expected Land Value (ELV)",
                           "Yearly Net Present Value (YNPV)",
                           "Internal Rate of Return (IRR)")
    
    
    # sensibilidade ####
    plot_data <- data.frame(x_axis = seq(sens_limits[1]/100, sens_limits[2]/100, 0.01) ) %>%
      dplyr::mutate(
        VPR = purrr::map_dbl(x_axis, ~sum(df[[revenue]] / (1 + .x)^df[[year]],na.rm=T)),
        VPC = purrr::map_dbl(x_axis, ~sum(df[[cost]] / (1 + .x)^df[[year]],na.rm=T)),
        VFR = purrr::map_dbl(x_axis, ~sum(df[[revenue]] * (1 + .x)^df[[year]],na.rm=T)),
        VFC = purrr::map_dbl(x_axis, ~sum(df[[cost]] * (1 + .x)^df[[year]],na.rm=T)),
        VPL = VPR - VPC,
        VET = (VFR - VFC)/((1+x_axis)^(length(df[[year]])-1)-1) ) %>% 
      dplyr::select(x_axis,NPV=VPL, ELV=VET) %>% 
      tidyr::gather("facet_var", "y_axis", NPV, ELV,factor_key = TRUE) %>% 
      dplyr::group_by(facet_var) %>% 
      dplyr::mutate(red_y_axis = ifelse(y_axis<0, y_axis, NA ),
                    red_y2_axis = any(y_axis < 0),
                    red_y2_axis = ifelse(red_y2_axis, 0, NA )
                    
      ) %>% as.data.frame()
    
    
    plot <- ggplot2::ggplot(plot_data, ggplot2::aes(x_axis, y_axis) ) + 
      ggplot2::geom_line(size=1)+{
        
        if(!all(is.na(plot_data$red_y_axis)))
          ggplot2::geom_line(ggplot2::aes(y=red_y_axis), color="red",size=1,na.rm=T )
        
      }+{
        
        if(!all(is.na(plot_data$red_y2_axis)))
          ggplot2::geom_line(ggplot2::aes(y=red_y2_axis), color="red",size=1,na.rm=T )  
        
      }+
      ggplot2::facet_wrap(~facet_var, scales="free_y",ncol=1,strip.position = "right") +
      ggplot2::scale_x_continuous(labels=scales::percent) +
      ggplot2::scale_y_continuous(labels=scales::dollar_format(big.mark=big_mark,decimal.mark = dec_mark,prefix=paste(prefix," ",sep="")))+
      ggplot2::labs(x="Rate",y=NULL) +
      ggthemes::theme_igray(base_family = "serif") +
      ggplot2::theme(
        axis.title       = ggplot2::element_text(size = 22), 
        axis.text        = ggplot2::element_text(size = 20),
        plot.title       = ggplot2::element_text(size = 22, face = "bold"),
        legend.title     = ggplot2::element_text(size = 20),
        legend.text      = ggplot2::element_text(size = 18),
        plot.caption     = ggplot2::element_text(size = 20),
        strip.text.y     = ggplot2::element_text(size = 18, face = "bold",angle=0),
        strip.background = ggplot2::element_blank() ) 
    
    # ####
    
    return(list(Sensibility=plot,Result=tab2_mod))
    
  }
  
  
}