date %in% (holidays_2013$date - 1L) ~ "day before holiday", date %in% (holidays_2013$date + 1L) ~ "day after holiday",
The exercise told you to focus on holidays rather than the days before or after holidays! you're always overthinking!
4 Matching Annotations
- Jul 2025
-
jrnold.github.io jrnold.github.io
-
- Jun 2025
-
jrnold.github.io jrnold.github.io
-
I used a list, not a character vector, since the class of an object can have multiple values. For example, the class of the time_hour column is POSIXct, POSIXt.
You think of everything man!
-
- May 2025
-
jrnold.github.io jrnold.github.io
-
Whether the mean is the best summary depends on what you are using it for :-), i.e. your objective.
- Whether the mean is the best summary metric depends on the distribution of the variable.
- If it is normal distribution, then mean is a good summary metric, otherwise, median is better.
- we can use function shapiro.test() to test the Normality of a variable
- since shapiro.test() limits the maximum length of a vector of 5000, we can subset 5000 elements of gss_cat$tvhours to perform this test.
gss_cat$tvhours %>% head(5000) %>% shapiro.test()
results: p-value < 0.01, which means the tvhours variable is not normal distribution, therefore, mean is not a good summary metric for this variable, we'd better use median.
-
- Mar 2025
-
jrnold.github.io jrnold.github.io
-
The question does not define a way to measure on-time record, so I will consider two metrics: proportion of flights not delayed or cancelled, and mean arrival delay.
This is just making a simple problem complicated.
worst on-time plane is just the plane with the biggest average arrival delay time.
-