Monthly Savings Calculator.R

#
# This is a Shiny web application. You can run the application by clicking
# the 'Run App' button above.
#
# Find out more about building applications with Shiny here:
#
#    http://shiny.rstudio.com/
#

# define the user interface
library(shiny)
library(shinythemes)
library(shinyMatrix)
library(ggplot2)
library(quantmod)
library(mice)
library(plotly)

ui <- fluidPage(
  
  theme = shinytheme("darkly"),
  
  titlePanel("Monthly Savings Calculator"),
  
  sidebarLayout(
    
    sidebarPanel(
      
      numericInput("initial_amount",
                   "Initial Amount",
                   value = 10000,
                   min = 0,
                   step = 1000),
      
      numericInput("final_amount",
                   "Desired Final Amount",
                   value = 100000,
                   min = 0,
                   step = 10000),
      
      numericInput("years",
                   "Years",
                   value = 5,
                   min = 1,
                   step = 1), # Changed step to 1
      
      numericInput("probability",
                   "Probability of Success",
                   value = 0.8,
                   min = 0,
                   max = 1,
                   step = 0.05),
      
      selectInput("emprical_distribution",
                  "Distribution from which to sample monthly Returns",
                  choices = c(
                    "Normal Distribution" = "ND",
                    "Portfolio Distribution" = "PD",
                    "Empirical Distribution" = "ED"
                  )),
      
      conditionalPanel(
        condition = "input.emprical_distribution == 'ED'",
        
        fileInput("file", "Choose CSV File",
                  multiple = FALSE,
                  accept = c("text/csv",
                             "text/comma-separated-values,text/plain",
                             ".csv")),
        
        checkboxInput("header", "Header", TRUE),
        
        radioButtons("sep", "Separator",
                     choices = c(Comma = ",",
                                 Semicolon = ";",
                                 Tab = "\t"),
                     selected = ","),
        
      ),
      
      conditionalPanel(
        condition = "input.emprical_distribution == 'ND'",
        
        numericInput("mean", "Arithmetic mean monthly Return", value = 0.0071, min = 0, step = 0.0001),
        numericInput("volatility", "Monthly Volatility", value = 0.0558, min = 0, step = 0.0001)
      ),
      
      conditionalPanel(
        condition = "input.emprical_distribution == 'PD'",
        
        matrixInput("input_matrix", "Portfolio",
                    value = matrix(data = c("SPYI.F", "BTC-EUR", 0.95, 0.05), nrow = 2, ncol = 2, dimnames = list(NULL, c("Ticker", "Weight"))),
                    rows = list(
                      extend = TRUE,
                      names = FALSE
                    ),
                    cols = list(
                      names = TRUE
                    )),
        
        checkboxInput("imputation", "Align periods by Imputation?", FALSE) 
        
      ),
      
      numericInput("simulations", "Simulations", value = 10000, min = 1),
      
      numericInput("seed", "Seed", value = 12345), # Provided a non-zero seed
      
      actionButton("submit", "Calculate")
      
    ),
    
    mainPanel(
      
      fluidRow(wellPanel(plotOutput("histogram"))),
      fluidRow(wellPanel(plotOutput("cdf"))),
      fluidRow(wellPanel(textOutput("text")))
      
    ) 
  )
)

server <- function(input, output) {
  
  observeEvent(input$submit, {
    
    set.seed(input$seed)
    
    # Number of months
    months <- input$years * 12
    
    # Matrix to store the returns
    gross_returns <- matrix(nrow = months, ncol = input$simulations)
    
    # Matrix to hold the amount of savings at each time step
    amount <- matrix(nrow = months + 1, ncol = input$simulations)
    amount[1, ] <- input$initial_amount
    
    if (input$emprical_distribution == "ED") {
      
      inFile <- input$file
      
      if (is.null(inFile)) # Added error handling for file input
        stop("Please upload a file.")
      
      data <- read.csv(inFile$datapath,
                       header = input$header,
                       sep = input$sep)
      
      names(data)[1] <- "Header"
      
      gross_returns <- replicate(n = input$simulations, expr = {
        sample(1 + data$Header, size = months, replace = TRUE)
      })
      
    } else if (input$emprical_distribution == "ND") {
      
      gross_returns <- replicate(n = input$simulations, expr = {
        1 + rnorm(n = months, mean = input$mean, sd = input$volatility)
      })
      
    } else {
      
      m <- input$input_matrix[!apply(input$input_matrix == "", 1, all),]
      
      if (is.null(dim(input$input_matrix[!apply(input$input_matrix == "", 1, all),]))) {
        m <- t(as.matrix(input$input_matrix[!apply(input$input_matrix == "", 1, all),]))
      } else {
        
      }
      
      getSymbols(m[, 1], src = "yahoo", from = "1900-01-01")
      
      assets <- c()
      
      for (i in 1:length(m[, 1])) {
        assets[[i]] <- monthlyReturn(get(m[i, 1])[, 6], leading = FALSE)
      }
      
      return_matrix <- do.call("merge.xts", assets)
      
      if (input$imputation == TRUE) {
        
        return_matrix <- mice(return_matrix,
                              m = max(round(sum(!complete.cases(return_matrix)) / nrow(return_matrix) * 100), 1),
                              maxit = length(m[, 1]))
        
        return_matrix <- complete(return_matrix)
        
      } else {
        
        return_matrix <- na.omit(return_matrix)
        
      }
      
      gross_returns <- replicate(n = input$simulations, expr = {
        sample(1 + (as.matrix(return_matrix) %*% as.matrix(as.numeric(m[, 2]))), size = months, replace = TRUE)
      })
      
    }
    
    min_savings <- 0
    max_savings <- input$final_amount %/% input$years
    
    while (min_savings < max_savings) {
      savings <- (min_savings + max_savings) %/% 2
      for (x in 1:months) {
        amount[x+1, ] <- (savings + amount[x, ]) * gross_returns[x, ]
      }
      if (sum(amount[(months + 1), ] > input$final_amount) / input$simulations >= input$probability) {
        max_savings <- savings
      } else {
        min_savings <- savings + 1
      }
    }
    
    output$histogram <- renderPlot({
      
      options(scipen = 999)
      ggplot(data = data.frame(amount[(months + 1), ]), aes(x = amount[(months + 1), ])) +
        geom_histogram(aes(y = after_stat(density)), bins = min(100, input$simulations), color = "black", fill = "black", alpha = 1) +
        geom_density(color = "red", alpha = 1, linewidth = 1) +
        geom_vline(xintercept = input$final_amount, color = "darkred", linewidth = 2) +
        ggtitle("Scenario Distribution") +
        labs(y = "Density", x = "Final Amount") +
        theme(text = element_text(size = 18, color = "black"),
              axis.text = element_text(size = 15, color = "black"),
              plot.title = element_text(hjust = 0.5))
    })
    
    output$cdf <- renderPlot({
      data <- data.frame(amount = amount[(months + 1), ])
      ggplot(data, aes(x = amount)) +
        stat_ecdf(geom = "step", color = "black", linewidth = 2) +
        geom_vline(xintercept = input$final_amount, color = "darkred", linewidth = 2) +
        geom_hline(yintercept = 1 - input$probability, color = "darkred", linewidth = 2) +
        ggtitle("Scenario CDF") +
        labs(x = "Final Amount", y = "Probability") +
        theme(text = element_text(size = 18, color = "black"),
              axis.text = element_text(size = 15, color = "black"),
              plot.title = element_text(hjust = 0.5))
    })
    
    
    output$text <- renderText({
      
      if (input$initial_amount + months * min_savings >= input$final_amount) {
        return(paste("You would have to invest as least ", min_savings, " each month in your risky securities portfolio, but it would be both safer and more capital efficient to simply save ", max(round((input$final_amount-input$initial_amount) / months, 0), 1), " each month without risk."))
      } else {
        return(paste("You have to invest at least ", min_savings, " each month in your risky securities portfolio. If you did not invest, but simply saved without risk, you would have to save ", as.integer((input$final_amount-input$initial_amount) / months), " each month. Accordingly, investing would be more capital efficient considering the given probability."))
      }
    })
  })
}

shinyApp(ui = ui, server = server)