# Load packages
library(tidyverse)
library(tidyquant)

1 Get stock prices and convert to returns

Ra <- c("NVDA", "AMD", "MSFT") %>%
    tq_get(get  = "stock.prices",
           from = "2022-01-01") %>%
    group_by(symbol) %>%
    tq_transmute(select     = adjusted, 
                 mutate_fun = periodReturn, 
                 period     = "monthly", 
                 col_rename = "Ra")
Ra
## # A tibble: 87 × 3
## # Groups:   symbol [3]
##    symbol date             Ra
##    <chr>  <date>        <dbl>
##  1 NVDA   2022-01-31 -0.187  
##  2 NVDA   2022-02-28 -0.00412
##  3 NVDA   2022-03-31  0.119  
##  4 NVDA   2022-04-29 -0.320  
##  5 NVDA   2022-05-31  0.00674
##  6 NVDA   2022-06-30 -0.188  
##  7 NVDA   2022-07-29  0.198  
##  8 NVDA   2022-08-31 -0.169  
##  9 NVDA   2022-09-30 -0.196  
## 10 NVDA   2022-10-31  0.112  
## # ℹ 77 more rows

2 Get baseline and convert to returns

Rb <- "^IXIC" %>%
    tq_get(get  = "stock.prices",
           from = "2022-01-01") %>%
    tq_transmute(select     = adjusted, 
                 mutate_fun = periodReturn, 
                 period     = "monthly", 
                 col_rename = "Rb")
Rb
## # A tibble: 29 × 2
##    date            Rb
##    <date>       <dbl>
##  1 2022-01-31 -0.101 
##  2 2022-02-28 -0.0343
##  3 2022-03-31  0.0341
##  4 2022-04-29 -0.133 
##  5 2022-05-31 -0.0205
##  6 2022-06-30 -0.0871
##  7 2022-07-29  0.123 
##  8 2022-08-31 -0.0464
##  9 2022-09-30 -0.105 
## 10 2022-10-31  0.0390
## # ℹ 19 more rows

3 Join the two data tables

RaRb <- left_join(Ra, Rb, by = c("date" = "date"))
RaRb
## # A tibble: 87 × 4
## # Groups:   symbol [3]
##    symbol date             Ra      Rb
##    <chr>  <date>        <dbl>   <dbl>
##  1 NVDA   2022-01-31 -0.187   -0.101 
##  2 NVDA   2022-02-28 -0.00412 -0.0343
##  3 NVDA   2022-03-31  0.119    0.0341
##  4 NVDA   2022-04-29 -0.320   -0.133 
##  5 NVDA   2022-05-31  0.00674 -0.0205
##  6 NVDA   2022-06-30 -0.188   -0.0871
##  7 NVDA   2022-07-29  0.198    0.123 
##  8 NVDA   2022-08-31 -0.169   -0.0464
##  9 NVDA   2022-09-30 -0.196   -0.105 
## 10 NVDA   2022-10-31  0.112    0.0390
## # ℹ 77 more rows

4 Calculate CAPM

RaRb_capm <- RaRb %>%
    tq_performance(Ra = Ra, 
                   Rb = Rb, 
                   performance_fun = table.CAPM)
RaRb_capm
## # A tibble: 3 × 13
## # Groups:   symbol [3]
##   symbol ActivePremium  Alpha AnnualizedAlpha  Beta `Beta-` `Beta+` Correlation
##   <chr>          <dbl>  <dbl>           <dbl> <dbl>   <dbl>   <dbl>       <dbl>
## 1 NVDA          0.705  0.0517          0.830  2.31    3.06    1.26        0.880
## 2 AMD           0.0262 0.0095          0.121  1.92    2.93    2.15        0.774
## 3 MSFT          0.0892 0.0078          0.0976 0.825   0.910   0.543       0.819
## # ℹ 5 more variables: `Correlationp-value` <dbl>, InformationRatio <dbl>,
## #   `R-squared` <dbl>, TrackingError <dbl>, TreynorRatio <dbl>

Wcich stock has a positively skewed distrobution of returns?

Both AMD and Microsoft have positive skeweness while Nvidia has a negative skewness.

RaRb_capm <- RaRb %>%
    tq_performance(Ra = Ra, 
                   Rb = NULL, 
                   performance_fun = skewness)
RaRb_capm
## # A tibble: 3 × 2
## # Groups:   symbol [3]
##   symbol skewness.1
##   <chr>       <dbl>
## 1 NVDA      -0.231 
## 2 AMD        0.0948
## 3 MSFT       0.148