According to the simulation results, MicroStrategy has had a positive trend with less volatility for the last ten years compared to AMAZON, which has had a negative trend with high volatility.

Source code:

library(tidyverse)
library(tidyquant)

df_port <- 
  tq_get(c("AMZN","MSTR")) %>% 
  group_by(symbol) %>%
  tq_transmute(select     = adjusted, 
               mutate_fun = periodReturn, 
               period     = "yearly", 
               type       = "arithmetic") %>% 
  mutate(date = floor_date(date, "year") %>% year(),
         symbol = case_when(
           symbol == "AMZN" ~ "AMAZON",
           symbol == "MSTR" ~ "MicroStrategy"
         )) %>% 
  group_by(symbol) %>% 
  slice_min(n = -1, order_by = date) %>% 
  ungroup() %>% 
  pivot_wider(names_from = symbol,
              values_from = yearly.returns) 

#Simulation
library(rsample)

set.seed(123)
port_intervals <- 
  reg_intervals(date ~ AMAZON + MicroStrategy, 
                data = df_port,
                type = "percentile",
                keep_reps = TRUE)



#Bootstrap confidence intervals
##https://juliasilge.com/blog/superbowl-conf-int/
port_intervals %>%
  mutate(
    term = str_remove(term, "TRUE"),
    term = fct_reorder(term, .estimate)
  ) %>%
  ggplot(aes(.estimate, term)) +
  geom_vline(xintercept = 0, size = 1.5, lty = 2, color = "gray80") +
  geom_errorbarh(aes(xmin = .lower, xmax = .upper),
                 size = 1.5, alpha = 0.5, color = "midnightblue"
  ) +
  geom_point(size = 3, color = "midnightblue") +
  labs(
    x = "",
    y = "",
    subtitle = "Trends in annual returns for Amazon and MicroStrategy\n(from 2015 to 2024)"
  ) +
  theme_minimal(base_family = "Roboto Slab", 
                base_size = 15) +
  theme(
    text = element_text(face = "bold")
  )