I am working with the R programming language.

I have this dataset:

myt = structure(list(name = c("Alice", "Bob", "Bob", "Charlie", "Diana", 
"Diana", "Eve", "Eve", "Frank", "Frank", "Frank", "Grace", "Grace", 
"Henry", "Henry", "Henry"), country = c("USA", "Canada", "Mexico", 
"UK", "France", "Germany", "USA", "Canada", "Mexico", "USA", 
"UK", "France", "Germany", "USA", "Canada", "Mexico"), date_arrived = structure(c(17897, 
18414, 18536, 18779, 18231, 18475, 18597, 18687, 18048, 18322, 
18567, 17897, 18762, 18383, 18489, 18597), class = "Date"), date_left = structure(c(18322, 
18506, 18659, 18840, 18444, 18567, 18659, 2932896, 18231, 18506, 
2932896, 18353, 18809, 18489, 18597, 18748), class = "Date")), class = "data.frame", row.names = c(NA, 
-16L))

I want to find out the following:

• For every person that has data between May-1-2020 and May-1-2021

• What percent of time did they spend in each country during this time period?

• If they have data between this whole time period, I use 365 as a denominator. If they do not, I use the total number of days they have, i.e. max(365, total days)

I tried to do this in stages.

First, I identified the period:

period_start <- as.Date("2020-05-01")
period_end <- as.Date("2021-05-01")
total_period_days <- as.numeric(period_end - period_start)

Then, I made some columns to calculate statistics and filter on people who only have data during this time:

myt$effective_arrival <- pmax(myt$date_arrived, period_start)
myt$effective_departure <- pmin(myt$date_left, period_end)

myt$num_days_in_range <- ifelse(
  myt$in_range == "yes",
  as.numeric(myt$effective_departure - myt$effective_arrival),
  0
)

myt_in_range <- myt[myt$in_range == "yes", ]

Then I did two aggregations:

days_by_country <- aggregate(
  num_days_in_range ~ name + country,
  data = myt_in_range,
  FUN = sum
)
names(days_by_country)[3] <- "days_in_country"

person_totals <- aggregate(
  days_in_country ~ name,
  data = days_by_country,
  FUN = sum
)
names(person_totals)[2] <- "total_days_present"

Finally, I did the merge and percent calculations:

person_totals$denominator <- pmin(person_totals$total_days_present, total_period_days)

final_results <- merge(days_by_country, person_totals, by = "name")

final_results$percentage <- round(
  100 * final_results$days_in_country / final_results$denominator, 
  2
)

final_results <- final_results[order(final_results$name, final_results$country), ]

final_results <- final_results[, c("name", "country", "days_in_country", 
                                   "total_days_present", "denominator", "percentage")]

Is there some way I can do this all at once in R using dplyr?

Thanks!