Ahmad Mousavi

Ahmad Mousavi

Assistant Prof. of Data Science,
Dept. of Math. and Stat.,
American University

Lab 10: Factors and Dates

DATA 412/612 · Week 10 · PDF · Source (.qmd)

Reshape a factor, then parse the WMATA ridership dates.

Work in a Quarto document. Do 10.1 and 10.2 first. Put wmata_ridership.csv in a data/ folder of your project (see the Data section of the course page).

library(tidyverse)
library(lubridate)

1. Factors

fc <- factor(c("D","C","C","A","A","C","D","A","C","C","A","A","C","A","C","B","C","A","C","B"))
  1. Add level "E" without adding any observations.
  2. Bar chart of fc by frequency, most frequent on the right.
  3. Put the levels in reverse alphabetical order.
  4. Remove the observations that are "C", then drop the unused level.
  5. Collapse "A" and "B" into "z".

2. Dates

Use data/wmata_ridership.csv.

  1. Parse Date. Add year, month, and wday columns with label = TRUE where it applies.
  2. Drop 2004.
  3. Boxplot of daily Total by weekday.
  4. How long is the span as a duration and as an interval?

3. Capital Bikeshare (sample)

Use data/capital_trips_sample.csv (a short extract of the 2016 trips file). Column names have spaces.

  1. glimpse() the file. Rename columns so they have no spaces.
  2. Parse start and end with mdy_hm().
  3. Compute trip duration from start and end. Compare it to duration_ms. Where do they disagree by about an hour?
  4. Those hour-scale gaps sit on 13 March 2016 (spring-forward). force_tz(..., tzone = "America/New_York") on both times, then recompute.
  5. Time from the first start to the last end, as a duration.
  6. Plot duration against hour of day.