Indonesia Stocks Analysis Dashboards

---
title: "Indonesia Stocks Analysis Dashboards"
output: 
  flexdashboard::flex_dashboard:
    vertical_layout: scroll
    theme: united
    source_code: embed
---




```{r setup, include=FALSE}
# Library Import

library(flexdashboard)
library(tidyverse)
library(highcharter)
library(gt)
library(htmltools)
library(viridis)
library(DT)
library(tidyquant)
library(plotly)
library(timetk)
library(lubridate)
library(zoo)
library(tidyr)

```

Menggunakan Library `quantmod` yang terhubung ke yahoo finance, Kita akan melakukan analisis terhadap 5 emiten saham di Indonesia. Yaitu: </br>
1. Adaro Energy Tbk. (ADRO) di bidang Energy </br>
2. Barito Pacific Tbk. (BRPT) di bidang Energy </br>
3. Indofood CBP Sukses Makmur Tbk. (ICBP) di bidang Produksi Pangan</br>
4. PT. Saratoga Investama Sedaya (SRTG) di bidang Investasi </br>
5. Sumber Alfaria Trijaya Tbk (AMRT) di bidang Perdagangan, Jasa dan Investasi

```{r, include = FALSE}
library(quantmod)
library(plotly)
library(PerformanceAnalytics)


tick = c('ADRO.JK', 'BRPT.JK', 'ICBP.JK','SRTG.JK', 'AMRT.JK')
ticklabel = c("Adaro Energy Tbk.", "Barito Pacific Renewable Energy", "Indofood CBP Sukses Makmur Tbk.", " PT. Saratoga Investama Sedaya ", "Sumber Alfaria Trijaya Tbk.")


stocksss = getSymbols(tick, from="2022-05-01", to="2024-05-01", src="yahoo", adjust=F)

stocksss
```

Visualisasi Harga Saham {data-orientation=columns}
=======================================================================
## Candlestick Chart {.tabset .tabset-fade data-height=1080}
-----------------------------------------------------------------------
###  Adaro Energy Tbk. {data-width=1920}

```{r}

priceadaro = tq_get('ADRO.JK',
                    from='2022-05-01' , 
                    to="2024-05-01", 
                    get='stock.prices')

priceadaro %>% plot_ly(x=~date, 
                       type = 'candlestick', 
                       open = ~open, 
                       close = ~close,
                       high = ~high,
                       low = ~low) %>%
  layout(title = "Chart Harga Saham per lembar Adaro Energy Tbk. Sejak Mei 2022", xaxis = list(rangeslider = list(visible= F)))


```

### Barito Pacific Tbk. {data-width=1920}

```{r}
pricebayan = tq_get('BRPT.JK',
                    from='2022-05-01' , 
                    to="2024-05-01", 
                    get='stock.prices')

pricebayan %>% plot_ly(x=~date, 
                       type = 'candlestick', 
                       open = ~open, 
                       close = ~close,
                       high = ~high,
                       low = ~low) %>%
  layout(title = "Chart Harga Saham per lembar Barito Pacific Tbk. Sejak Mei 2022", xaxis = list(rangeslider = list(visible= F)))
```

### Indofood CBP Sukses Makmur Tbk. {data-width=1920}
```{r}
priceICBP = tq_get('ICBP.JK',
                    from='2022-05-01' , 
                    to="2024-05-01", 
                    get='stock.prices')

priceICBP %>% plot_ly(x=~date, 
                       type = 'candlestick', 
                       open = ~open, 
                       close = ~close,
                       high = ~high,
                       low = ~low) %>%
  layout(title = "Chart Harga Saham per lembar Indofood CBP Sukses Makmur Tbk. Sejak Mei 2022", xaxis = list(rangeslider = list(visible= F)))
```
### PT Saratoga Investama Sedaya Tbk. {data-width=1920}

```{r}
priceSRTG = tq_get('SRTG.JK',
                    from='2022-05-01' , 
                    to="2024-05-01", 
                    get='stock.prices')

priceSRTG %>% plot_ly(x=~date, 
                       type = 'candlestick', 
                       open = ~open, 
                       close = ~close,
                       high = ~high,
                       low = ~low) %>%
  layout(title = "Chart Harga Saham per lembar Sarataga Investama Sedaya Tbk. Sejak Mei 2022", xaxis = list(rangeslider = list(visible= F)))
```

### Sumber Alfaria Trijaya Tbk {data-width=1920}

```{r}
priceAMRT = tq_get('AMRT.JK',
                    from='2022-05-01' , 
                    to="2024-05-01", 
                    get='stock.prices')

priceAMRT %>% plot_ly(x=~date, 
                       type = 'candlestick',
                       open = ~open, 
                       close = ~close,
                       high = ~high,
                       low = ~low) %>%
  layout(title = "Chart Harga Saham per lembar Alfaria Trijaya Tbk. Sejak Mei 2022", xaxis = list(rangeslider = list(visible= F)))

```

Visualisasi daily Return {data-orientation=columns}
=======================================================================
## Daily Return {.tabset .tabset-fade data-height=1000}
-----------------------------------------------------------------------

```{r, include=FALSE}
##  Calculate daily returns  and Combine  (B&C)
closing = NULL
for (stock in tick) { closing = cbind(closing, Cl(get(stock)))}

return = Return.calculate(closing)
return = data.frame(return)
head (return)
```

### ADRO Return Plot
```{r}

 plot_ly(x = priceadaro$date, y = ~as.numeric(return$ADRO.JK.Close), type = 'scatter', mode = 'lines', name = 'Actual') %>%
  layout(title = 'Actual vs. ARIMA Forecast',
         xaxis = list(title = 'Time'),
         yaxis = list(title = 'Return'))

```

#### ADARO Average Returns
```{r}
cat("ADRO Average Daily Returns:", mean(na.omit(return$ADRO.JK.Close)), "\n")
```

### BRPT Return Plot

```{r}
 plot_ly(x = pricebayan$date, y = ~as.numeric(return$BRPT.JK.Close), type = 'scatter', mode = 'lines', name = 'Actual') %>%
  layout(title = 'Actual vs. ARIMA Forecast',
         xaxis = list(title = 'Time'),
         yaxis = list(title = 'Return'))

```

```{r}
cat("BRPT Average Daily Returns:", mean(na.omit(return$BRPT.JK.Close)), "\n")
```

### ICBP Retrun Plot

```{r}
 plot_ly(x = priceICBP$date, y = ~as.numeric(return$ICBP.JK.Close), type = 'scatter', mode = 'lines', name = 'Actual') %>%
  layout(title = 'Actual vs. ARIMA Forecast',
         xaxis = list(title = 'Time'),
         yaxis = list(title = 'Return'))

```

```{r}
cat("ICBP Average Daily Returns:", mean(na.omit(return$ICBP.JK.Close)), "\n")
```


### Daily Return of Saratgo

```{r}
 plot_ly(x = priceSRTG$date, y = ~as.numeric(return$SRTG.JK.Close), type = 'scatter', mode = 'lines', name = 'Actual') %>%
  layout(title = 'Actual vs. ARIMA Forecast',
         xaxis = list(title = 'Time'),
         yaxis = list(title = 'Return'))

```


```{r}
cat("SRTG Average Daily Returns:", mean(na.omit(return$SRTG.JK.Close)), "\n")
```

### Daily Return of AMRT

```{r}
 plot_ly(x = priceAMRT$date, y = ~as.numeric(return$AMRT.JK.Close), type = 'scatter', mode = 'lines', name = 'Actual') %>%
  layout(title = 'Actual vs. ARIMA Forecast',
         xaxis = list(title = 'Time'),
         yaxis = list(title = 'Return'))

```

```{r}
cat("AMRT Average Daily Returns:", mean(na.omit(return$AMRT.JK.Close)), "\n")
```


Correlation Heatmap {data-orientation=columns}
=======================================================================
## Correlation Analysis {.tabset .tabset-fade data-height=640}
-----------------------------------------------------------------------



```{r}
return = na.omit(return)
corstock = cor(na.omit(return))

mask = matrix(F, nrow=nrow(corstock), ncol=ncol(corstock))
mask[lower.tri(mask)] = T

heatmap = plot_ly(z=corstock, x=ticklabel, y=ticklabel, type="heatmap")
heatmap = heatmap %>%
  layout(title = "Correlation Heatmap of Indonesian Stocks Return Price", 
         xaxis = list(title=""), 
         yaxis=list(title=""),
         colorscale= "Inferno",
         margin = list(l=60, r=80, t=100, b=60))

heatmap
```


Portofolio Return and Volatility {data-orientation=columns}
=======================================================================
## Portofolio Return  {.tabset .tabset-fade data-height=640}
-----------------------------------------------------------------------


```{r}

 num_stocks <- length(closing)
  weights <- rep(1/num_stocks, num_stocks)
  
  portfolio_returns <- rowSums(closing * weights)
    portfolio_sd <- sqrt(var(portfolio_returns))
    
cat("Portfolio Returns:", mean(portfolio_returns), "\n")
cat("Portfolio Volatility:", portfolio_sd, "\n")

```

If a portfolio has a return of 7.85%, it means that over a specific period (in this case 2 years), the value of the portfolio has increased by 7.85% compared to its initial value. With the standard Deviation of 0.34%. Which brings us to the growth interval of 7.5 - 8.2%. 


Diversification Ratio {data-orientation=columns}
=======================================================================
## Diversification Ratio  {.tabset .tabset-fade data-height=640}
-----------------------------------------------------------------------

### Diversification Ratio for Each Stocks

```{r}
return = na.omit(return)
column_mean = apply(return, 2, mean, na.rm=T)
column_sd <- apply(return, 2, sd, na.rm = TRUE)

column_sd = as.data.frame(column_sd)

fig <- plot_ly(x = tick, y = column_sd$column_sd, type = 'bar',
        marker = list(color = 'rgb(158,202,225)',
                      line = list(color = 'rgb(8,48,107)',
                                  width = 1.5)))
fig <- fig %>% layout(title = "Diversification ratio",
         xaxis = list(title = ""),
         yaxis = list(title = ""))

fig
```


### Diversification Joint Ratio 

```{r}

calculate_diversification_ratio <- function(portfolio_sd, column_sd) {
  weighted_avg_individual_sd <- sum(column_sd) / length(column_sd)
  
  # Calculate diversification ratio
  diversification_ratio <- weighted_avg_individual_sd / portfolio_sd
  
  return(diversification_ratio)
}

diversification_ratio <- calculate_diversification_ratio(portfolio_sd, column_sd)
cat("Diversification Ratio:", diversification_ratio, "\n")
```
Optimal Weights  {data-orientation=columns}
=======================================================================
## Optimal Weights for Each Stocks {.tabset .tabset-fade data-height=640}
-----------------------------------------------------------------------

### Optimal Distribution 

```{r}
library(quadprog)
# Example data (replace with your actual data)
expected_returns <- c(7.854, 7.854, 7.854, 7.854, 7.854 )  # Expected returns for assets
cov_matrix <- matrix(corstock, nrow=5)  # Covariance matrix

library(Matrix)
pd_D_mat <- nearPD(cov_matrix)

# Use the positive definite matrix for optimization
D_mat <- as.matrix(pd_D_mat$mat)
d_vec <- rep(0, length(expected_returns))
A_mat <- cbind(rep(1, length(expected_returns)), diag(length(expected_returns)))
b_vec <- c(1, d_vec)

# Solve quadratic programming problem
library(quadprog)
output <- solve.QP(Dmat = D_mat, dvec = d_vec, Amat = A_mat, bvec = b_vec, meq = 1)

# Optimal portfolio weights
optimal_weights <- output$solution

pie <- plot_ly(labels = ticklabel, values = optimal_weights, type = "pie")

pie <- pie %>% layout(title = "Optimized Portfolio Weight", xaxis = list(title = ""), yaxis = list(title = ""))

pie
```

Stock Forecasting  {data-orientation=columns}
=======================================================================
## Stocks Forecasting using ARIMA {.tabset .tabset-fade data-height=640}
-----------------------------------------------------------------------

### Adaro Energy Tbk. 

```{r}
library(tseries)
library(forecast)

arimaadaro = auto.arima(priceadaro$close)
fcadaro = forecast(arimaadaro)

start = ymd("2024-05-02")
end = ymd("2024-05-11")
fdate = seq(start, end, by = "day")


plot_time_series <- plot_ly(x =  as.Date(priceadaro$date), y = ~as.numeric(priceadaro$close), type = 'scatter', mode = 'lines', name = 'Actual') %>%
  add_trace( x = as.Date(fdate),y = fcadaro$mean, mode = 'lines', name = 'Predicted (ARIMA)') %>%
  layout(title = 'Actual vs. ARIMA Forecast',
         xaxis = list(title = 'Time'),
         yaxis = list(title = 'Stock Price'))

plot_time_series
```

###  Barito Pacific Energy Tbk. 

```{r}
library(tseries)
library(forecast)

arimabayan = auto.arima(pricebayan$close)
fcbayan = forecast(arimabayan)

start = ymd("2024-05-02")
end = ymd("2024-05-11")
fdate = seq(start, end, by = "day")


plot_time_series <- plot_ly(x =  as.Date(pricebayan$date), y = ~as.numeric(pricebayan$close), type = 'scatter', mode = 'lines', name = 'Actual') %>%
  add_trace( x = as.Date(fdate),y = fcbayan$mean, mode = 'lines', name = 'Predicted (ARIMA)') %>%
  layout(title = 'Actual vs. ARIMA Forecast',
         xaxis = list(title = 'Time'),
         yaxis = list(title = 'Stock Price'))

plot_time_series
```

### Indofood CBP Sukses Makmur

```{r}
library(tseries)
library(forecast)

arimaicbp = auto.arima(priceICBP$close)
fcicbp = forecast(arimaicbp)

start = ymd("2024-05-02")
end = ymd("2024-05-11")
fdate = seq(start, end, by = "day")


plot_time_series <- plot_ly(x =  as.Date(priceICBP$date), y = ~as.numeric(priceICBP$close), type = 'scatter', mode = 'lines', name = 'Actual') %>%
  add_trace( x = as.Date(fdate),y = fcicbp$mean, mode = 'lines', name = 'Predicted (ARIMA)') %>%
  layout(title = 'Actual vs. ARIMA Forecast',
         xaxis = list(title = 'Time'),
         yaxis = list(title = 'Stock Price'))

plot_time_series
```

### Saratoga Investama

```{r}
library(tseries)
library(forecast)

arimasrtg = auto.arima(priceSRTG$close)
fcsrtg = forecast(arimasrtg)

start = ymd("2024-05-02")
end = ymd("2024-05-11")
fdate = seq(start, end, by = "day")


plot_time_series <- plot_ly(x =  as.Date(priceSRTG$date), y = ~as.numeric(priceSRTG$close), type = 'scatter', mode = 'lines', name = 'Actual') %>%
  add_trace( x = as.Date(fdate),y = fcsrtg$mean, mode = 'lines', name = 'Predicted (ARIMA)') %>%
  layout(title = 'Actual vs. ARIMA Forecast',
         xaxis = list(title = 'Time'),
         yaxis = list(title = 'Stock Price'))

plot_time_series
```

### Alfamart Trijaya Tbk. 

```{r}
library(tseries)
library(forecast)

arimaamrt = auto.arima(priceAMRT$close)
fcamrt = forecast(arimaamrt)

start = ymd("2024-05-02")
end = ymd("2024-05-11")
fdate = seq(start, end, by = "day")


plot_time_series <- plot_ly(x =  as.Date(priceAMRT$date), y = ~as.numeric(priceAMRT$close), type = 'scatter', mode = 'lines', name = 'Actual') %>%
  add_trace( x = as.Date(fdate),y = fcamrt$mean, mode = 'lines', name = 'Predicted (ARIMA)') %>%
  layout(title = 'Actual vs. ARIMA Forecast',
         xaxis = list(title = 'Time'),
         yaxis = list(title = 'Stock Price'))

plot_time_series
```


Value At Risk  {data-orientation=columns}
=======================================================================
## Value at Risk {.tabset .tabset-fade data-height=640}
-----------------------------------------------------------------------

### Portofolio VAR

```{r}
# Define portfolio weights
weights <- optimal_weights

# Calculate covariance matrix
cov_matrix <- cov(return)

# Calculate portfolio variance
port_variance <- t(weights) %*% cov_matrix %*% weights

# Calculate portfolio standard deviation
port_sd <- sqrt(port_variance)

# Define confidence level and time horizon
confidence_level <- 0.95
time_horizon <- 1  # 1 day

# Calculate VaR
VaR <- qnorm(confidence_level) * port_sd * sqrt(time_horizon)
VaR
```

Dengan nilai 2% artinya, hanya sekitar 2% daripada total aset portofolio kita yang dipertaruhkan, apakah akan hilang atau bertambah. 


### VAR By Copula

```{r}
library(copula)
library(fitdistrplus)
library(plotly)

# Fit marginal distributions to the returns
fit_ADRO <- fitdist(as.vector(return[, 1]), "norm")
fit_BRPT <- fitdist(as.vector(return[, 2]), "norm")
fit_ICBP <- fitdist(as.vector(return[, 3]), "norm")
fit_SRTG <- fitdist(as.vector(return[, 4]), "norm")
fit_AMRT <- fitdist(as.vector(return[, 5]), "norm")

# Transform returns to uniform using the CDF of the fitted distributions
u_ADRO <- pnorm(return[, 1], mean = fit_ADRO$estimate["mean"], sd = fit_ADRO$estimate["sd"])
u_BRPT <- pnorm(return[, 2], mean = fit_BRPT$estimate["mean"], sd = fit_BRPT$estimate["sd"])
u_ICBP <- pnorm(return[, 3], mean = fit_ICBP$estimate["mean"], sd = fit_ICBP$estimate["sd"])
u_SRTG <- pnorm(return[, 4], mean = fit_SRTG$estimate["mean"], sd = fit_SRTG$estimate["sd"])
u_AMRT <- pnorm(return[, 5], mean = fit_AMRT$estimate["mean"], sd = fit_AMRT$estimate["sd"])
# Combine the uniform margins
u_matrix <- cbind(u_ADRO, u_BRPT, u_ICBP, u_SRTG, u_AMRT)

# Fit a Gaussian copula to the uniform data
cop <- fitCopula(normalCopula(dim = 5), u_matrix, method = "ml")

# Simulate from the fitted copula
set.seed(123)
n_sim <- 1000
u_sim <- rCopula(n_sim, cop@copula)

# Transform back to the original margins
sim_ADRO <- qnorm(u_sim[,1], mean = fit_ADRO$estimate["mean"], sd = fit_ADRO$estimate["sd"])
sim_BRPT <- qnorm(u_sim[,2], mean = fit_BRPT$estimate["mean"], sd = fit_BRPT$estimate["sd"])
sim_ICBP <- qnorm(u_sim[,3], mean = fit_ICBP$estimate["mean"], sd = fit_ICBP$estimate["sd"])
sim_SRTG <- qnorm(u_sim[,4], mean = fit_SRTG$estimate["mean"], sd = fit_SRTG$estimate["sd"])
sim_AMRT <- qnorm(u_sim[,5], mean = fit_AMRT$estimate["mean"], sd = fit_AMRT$estimate["sd"])

# Combine simulated returns
sim_returns <- cbind(sim_ADRO, sim_BRPT, sim_ICBP, sim_SRTG, sim_AMRT)
colnames(sim_returns) <- ticklabel

# Define portfolio weights (equal weighting as an example)
weights <- optimal_weights

# Calculate portfolio returns from simulated data
sim_port_returns <- rowSums(sim_returns * weights)

# Define confidence level
confidence_level <- 0.95

# Calculate VaR
VaR_copula <- quantile(sim_port_returns, probs = 1 - confidence_level)
plot_ly(x = ~sim_port_returns, type = "histogram", name = "Simulated Portfolio Returns") %>%
  layout(
    shapes = list(
      list(
        type = "line",
        x0 = VaR_copula, x1 = VaR_copula,
        y0 = 0, y1 = max(hist(sim_port_returns, plot=FALSE)$density),
        line = list(color = "red", width = 2),
        xref = "x", yref = "y"
      )
    ),
    title = "Simulated Portfolio Returns with VaR",
    xaxis = list(title = "Returns"),
    yaxis = list(title = "Frequency")
  )
```

  Menggunakan Copula Didapatkan Juga Bahwa VAR sekitar -2.2%