# Load packages

# Core
library(tidyverse)
library(tidyquant)

Goal

Visualize expected returns and risk to make it easier to compare the performance of multiple assets and portfolios.

Choose your stocks.

from 2012-12-31 to 2017-12-31

1 Import stock prices

# Choose stocks


symbols <- c("MELI", "SHOP", "TTD", "AFL", "NVDA")

prices <- tq_get(x = symbols,
                 get = "stock.prices",
                 from = "2012-12-31",
                 to = "2017-12-31")

2 Convert prices to returns (monthly)

asset_returns_tbl <- prices %>%
    # Calculate monthly returns
    group_by(symbol) %>%
    tq_transmute(select = adjusted,
                 mutate_fun = periodReturn,
                 period = "monthly",
                 type = "log") %>%
    slice(-1) %>%
    ungroup() %>%
    # remane
set_names(c("asset", "date", "returns"))

3 Assign a weight to each asset (change the weigting scheme)

symbols <- asset_returns_tbl %>% distinct(asset) %>% pull()
w <- c(0.30,
       0.15,
       0.10,
       0.20,
       0.20)
w_tbl <- tibble(symbols, w)

4 Build a portfolio

portfolio_returns_tbl <- asset_returns_tbl %>%
    
    tq_portfolio(assets_col   = asset,
                 returns_col  = returns,
                 weights      = w_tbl,
                 col_rename   = "returns",
                 rebalance_on = "months")
portfolio_returns_tbl
## # A tibble: 60 × 2
##    date         returns
##    <date>         <dbl>
##  1 2013-01-31  0.0174  
##  2 2013-02-28 -0.0170  
##  3 2013-03-28  0.0318  
##  4 2013-04-30  0.0268  
##  5 2013-05-31  0.0338  
##  6 2013-06-28  0.000740
##  7 2013-07-31  0.0335  
##  8 2013-08-30 -0.0136  
##  9 2013-09-30  0.0458  
## 10 2013-10-31  0.0114  
## # … with 50 more rows

5 Compute Standard Deviation

portfolio_sd_tidyquant_builtin_percent <- portfolio_returns_tbl %>%
    
    tq_performance(Ra = returns,
                   performance_fun = table.Stats) %>%
    
    select(Stdev) %>%
    mutate(tq_sd = round(Stdev, 3))

portfolio_sd_tidyquant_builtin_percent
## # A tibble: 1 × 2
##    Stdev tq_sd
##    <dbl> <dbl>
## 1 0.0444 0.044
# Mean of portfolio returns
portfolio_mean_tidyqaunt_builtin_percent <-
mean(portfolio_returns_tbl$returns)
    
portfolio_mean_tidyqaunt_builtin_percent
## [1] 0.01722598

6 Plot: Expected Returns versus Risk

# Expected Returns VS risk
sd_mean_tbl <- asset_returns_tbl %>%
    
    group_by(asset) %>%
    tq_performance(Ra = returns,
                   performance_fun = table.Stats) %>%
    select(Mean = ArithmeticMean, Stdev) %>%
    ungroup() %>%
    mutate(Stdev = Stdev * 100,
           Mean = Mean *100) %>%
    
    # Add portfolio sd
    add_row(tibble(asset = "portfolio",
                   Mean = portfolio_mean_tidyqaunt_builtin_percent * 100,
                   Stdev = portfolio_sd_tidyquant_builtin_percent$tq_sd)) 
sd_mean_tbl
## # A tibble: 6 × 3
##   asset      Mean  Stdev
##   <chr>     <dbl>  <dbl>
## 1 AFL        1.04  4.1  
## 2 MELI       2.35 10.7  
## 3 NVDA       4.71  8.81 
## 4 SHOP       4.23 13.4  
## 5 TTD        2.99 17.7  
## 6 portfolio  1.72  0.044
sd_mean_tbl %>%
    
    ggplot(aes(x = Stdev, y = Mean, color = asset)) +
    geom_point() +
    ggrepel::geom_label_repel(aes(label = asset))

### 24 Months Rolling Volatility

rolling_sd_tbl <- portfolio_returns_tbl %>%
    
    tq_mutate(select = returns,
              mutate_fun = rollapply,
              width = 24,
              FUN = sd,
              col_rename = "rolling_sd") %>%
    
    na.omit() %>%
    select(date, rolling_sd)

rolling_sd_tbl
## # A tibble: 37 × 2
##    date       rolling_sd
##    <date>          <dbl>
##  1 2014-12-31     0.0267
##  2 2015-01-30     0.0276
##  3 2015-02-27     0.0286
##  4 2015-03-31     0.0283
##  5 2015-04-30     0.0283
##  6 2015-05-29     0.0277
##  7 2015-06-30     0.0281
##  8 2015-07-31     0.0276
##  9 2015-08-31     0.0349
## 10 2015-09-30     0.0341
## # … with 27 more rows
rolling_sd_tbl %>%
    
    ggplot(aes(x = date, y = rolling_sd)) +
    geom_line(color = "cornflowerblue") +
    
    # Formatting
    scale_y_continuous(labels = scales::percent_format()) +
     
    # Labeling
    labs(x = NULL,
         y = NULL,
         title = "24-Month Rolling Volatility")

How should you expect your portfolio to perform relative to its assets in the portfolio? Would you invest all your money in any of the individual stocks instead of the portfolio? Discuss both in terms of expected return and risk.
## AFL has the lowest returns but it is also the least risky relative to the rest of the portfolio. On the other hand NVDA and MELI have similar risk but the returns of NVDA are higher relative to MELI. Personally I would put most of my money into NVDA and AFL relative to the rest of the portfolio. SHOP and TTD are higher risk with high returns relative to AFL and MELI.