Introduction

This application can be used by investment companies and individuals with a stock portfolio.
It will be useful to them for the analysis and optimization of their portfolio.
Thanks to this application, the Portfolio Managers will be able to effectively analyze the data of the assets in their charge and thus make informed decisions concerning their investments.
In addition, they will be able to better advise their current and future customers.
It is a high-performance, dynamic, easy-to-use application that has all the necessary features for portfolio creation, analysis and optimization.

Play with my Shiny web app at this link

App features

 Possibility for the user to choose any stock market index between the CAC40, the S&P500, the DAX, the FTSE100 and the Nikkei225;

 Possibility for the user to choose the number of assets constituting the portfolio as well as the weight of each asset;

 Retrieval of the daily closing prices of the Shares over a given period and possibility of saving them in CSV format;

 Display of statistics (average, median, quartiles, etc.) on closing prices;

 Closing price visualization charts (Histograms, line chart of each asset);

 Calculation of the Daily Yields of each Share and possibility of saving them in CSV format;

 Display of statistics (mean, median, quartiles, etc.) of Returns;

 Visualization of Yields (Histograms and Line Graphs);

 Calculation and visualization of correlations (correlation between each pair of assets);

 Visualization graphs of the evolution of the Beginning of Period and End of Period parameters;

 Aggregation of daily yields to obtain average annual, quarterly, monthly or weekly yields and possibility of saving them in CSV format;

 Calculation of moving annualized average returns, moving annualized volatilities, moving annualized sharpe ratios and visualization of their evolution (the user must be able to choose the size of the moving window. This size will be expressed in number of months);

 Calculation and Visualization of Drawdowns (potential losses);

 Portfolio optimization by different methods (Mean-Variance Efficient and Imposing Constraints);

 Optimization evaluation.

How i make this app?

Import the required libraries:

# Libraries
library(PerformanceAnalytics)
library(quantmod)
library(tseries)

Import the data from Yahoo Finance:

# Importation des données de plusieurs titres

stocks <- new.env()

stocks_names <- c("GOOG", "AMZN", "AAPL", "MSFT", "IBM")

getSymbols(stocks_names, 
           env = stocks, 
           from = as.Date('2015-01-01'), 
           to = as.Date('2020-01-01'))

# Extraction des prix de clôture de chaque titre pour en former un seul ensemble de données au format xts

close_prices <- do.call(merge, lapply(stocks, Cl))

# Affichage des premières observations

head(close_prices)

Statistic:

# Statistiques des prix de clôture

summary(close_prices)

Histogramm of closed prices for all stocks:

# Histogrammes des prix de tous les stocks

par(mfrow = c(2, 3))

for (i in 1:ncol(close_prices)) {
  chart.Histogram(close_prices[, i], 
                  xlab = 'Prix de clôture')
}

Evolution of closing prices over time:

# Prix de clôture au cours du temps

plot.zoo(close_prices, 
         main = 'Evolution des prix de clôture', 
         col = 1:ncol(close_prices))

Calculation of daily returns:

# Calcul des rendements journaliers de chaque stock

stocks_daily_returns <- Return.calculate(close_prices)

# Suppression de la première ligne

stocks_daily_returns <- stocks_daily_returns[-1]

Creating a portfolio and calculating its returns:

# Calcul des rendements du portefeuille à pondération égale

portfolio <- Return.portfolio(stocks_daily_returns, 
                             weights = c(1/5, 1/5, 1/5, 1/5, 1/5), 
                             rebalance_on = NA, 
                             verbose = TRUE)

# Fusion des rendements de stocks et du portefeuille

returns <- merge(stocks_daily_returns, portfolio$returns)

Statistics and Histograms of Returns:

# Histogramme des rendements des stocks et de ceux du portefeuille

par(mfrow = c(2, 3))

for (i in 1:ncol(returns)) {
  
  chart.Histogram(returns[, i])
  
}

Evolution of Portfolio’returns:

# Evolution des rendements des stocks et ceux du portefeuille

plot.zoo(returns, 
         main = 'Rendements des Stocks et du Portefeuille', 
         col = 1:ncol(returns) + 1)

Beginning of Period (BOP) et End of Period (EOP):

# Evolution des BOP

plot.zoo(portfolio$BOP.Weight, 
         main = 'BOP.Weight', 
         col = 1:ncol(stocks_daily_returns))

# Evolution des EOP

plot.zoo(portfolio$EOP.Weight, 
         main = 'EOP.Weight', 
         col = 1:ncol(stocks_daily_returns))

Aggregate average returns:

# Rendements moyens mensuels du portefeuille

pf_monthly_returns <- apply.monthly(portfolio$returns, mean)

# Evolution au cours du temps des rendements moyens mensuels

plot(pf_monthly_returns, 
     main = "Rendements moyens mensuels du portefeuille")

addLegend("topleft", on=1, lty = 1, 
          legend.names = colnames(pf_monthly_returns))

Correlation between asset returns:

# Corrélation entre les rendements des actifs

chart.Correlation(stocks_daily_returns, 
                  histogram = FALSE)

Rolling annualized average, standard deviation and Sharpe ratio of returns:

# Rendements moyens annualisés mobiles sur 12 mois

chart.RollingPerformance(R = pf_monthly_returns, 
                         width = 12,
                         FUN = "Return.annualized")

# Ecart-type des Rendements annualisés mobiles sur 12 mois

chart.RollingPerformance(R = pf_monthly_returns, 
                         width = 12,
                         FUN = "StdDev.annualized")

# Ratio de sharpe annualisés mobiles sur 12 mois

chart.RollingPerformance(R = pf_monthly_returns, 
                         width = 12,
                         FUN = "SharpeRatio.annualized")

Calculate of Drawdowns:

# Pertes potentielles du portefeuille

chart.Drawdown(pf_monthly_returns, 
               main = "Evolution des pertes potentielles du portefeuille")

Efficient mean-variance portfolio:

# Portefeuille efficace de moyenne-variance

opt1 <- portfolio.optim(
  apply.monthly(stocks_daily_returns, mean)
  )

# Poids optimaux du portefeuille de moyenne-variance

w <- opt1$pw

names(w) <- colnames(stocks_daily_returns)

print(w)

# Diagramme à barre montrant les poids optimaux

barplot(w)

R SHINY WEB APP FOR FINANCIAL PORTFOLIO ANALYSIS

My Reproductible Pitch presentation

Introduction

App features

How i make this app?