Tidyverse Book

Lubridate Package:

Makes working with date-time data easier. Base R commands for date-times are generally unintuitive not robust. According to Hadley Wickham, “lubridate has a consistent, memorable syntax, that makes working with dates fun instead of frustrating.” For additional information use help(package = lubridate) to bring up an overview of the package and its functions.

Load Packages:

library(readr)
library(lubridate)
library(dplyr)
library(knitr)

Load data from the FiveThirtyEight article “Some People Are Too Superstitious To Have A Baby On Friday The 13th”

Since it is fairly clean data, I will create additional columns to showcase lubridate functionality below
I will use the now(), hour(), minute() and second() functions to add addditional metadata to the dataframe. Please see definitions below
Once the new columns are added, I will remove the original clean columns

US_births_2000 <- read_csv("https://raw.githubusercontent.com/fivethirtyeight/data/master/births/US_births_1994-2003_CDC_NCHS.csv")%>%
                filter(year == 2000)%>%
                mutate(year = as.character(year),
                       month = as.character(month),
                       date_of_month = as.character(date_of_month)) %>%
               mutate(ymd.raw = paste(year, month, date_of_month, sep = '-'))%>%
               mutate(mdy.raw = paste(month,date_of_month,year, sep = '-'))%>%
               mutate(dmy.raw = paste(date_of_month,month, year,sep = '-'))%>%
              mutate(ymd.hms.raw = paste(year,month,date_of_month, hour(now()), minute(now()), second(now()),sep = '-'))%>%
              select(-c(year, month, date_of_month))

kable(head(US_births_2000,10),  caption = "2000 Birth Data")

2000 Birth Data
day_of_week	births	ymd.raw	mdy.raw	dmy.raw	ymd.hms.raw
6	8843	2000-1-1	1-1-2000	1-1-2000	2000-1-1-12-49-10.9784529209137
7	7816	2000-1-2	1-2-2000	2-1-2000	2000-1-2-12-49-10.9784529209137
1	11123	2000-1-3	1-3-2000	3-1-2000	2000-1-3-12-49-10.9784529209137
2	12703	2000-1-4	1-4-2000	4-1-2000	2000-1-4-12-49-10.9784529209137
3	12240	2000-1-5	1-5-2000	5-1-2000	2000-1-5-12-49-10.9784529209137
4	12260	2000-1-6	1-6-2000	6-1-2000	2000-1-6-12-49-10.9784529209137
5	12280	2000-1-7	1-7-2000	7-1-2000	2000-1-7-12-49-10.9784529209137
6	8750	2000-1-8	1-8-2000	8-1-2000	2000-1-8-12-49-10.9784529209137
7	7736	2000-1-9	1-9-2000	9-1-2000	2000-1-9-12-49-10.9784529209137
1	11418	2000-1-10	1-10-2000	10-1-2000	2000-1-10-12-49-10.9784529209137

Functionality Overview:

Parsing Date Information:

ymd(): parses in order - year, month, date. I will apply to the “ymd.raw” column and save the results as “ymd.clean”

cleaned_date_df <- US_births_2000%>%
                  mutate(ymd.clean = ymd(ymd.raw),
                         ymd.type = class(ymd.clean))%>%
                  select(-c(ymd.raw))
kable(head(cleaned_date_df,10),  caption = "ymd() results")

ymd() results
day_of_week	births	mdy.raw	dmy.raw	ymd.hms.raw	ymd.clean	ymd.type
6	8843	1-1-2000	1-1-2000	2000-1-1-12-49-10.9784529209137	2000-01-01	Date
7	7816	1-2-2000	2-1-2000	2000-1-2-12-49-10.9784529209137	2000-01-02	Date
1	11123	1-3-2000	3-1-2000	2000-1-3-12-49-10.9784529209137	2000-01-03	Date
2	12703	1-4-2000	4-1-2000	2000-1-4-12-49-10.9784529209137	2000-01-04	Date
3	12240	1-5-2000	5-1-2000	2000-1-5-12-49-10.9784529209137	2000-01-05	Date
4	12260	1-6-2000	6-1-2000	2000-1-6-12-49-10.9784529209137	2000-01-06	Date
5	12280	1-7-2000	7-1-2000	2000-1-7-12-49-10.9784529209137	2000-01-07	Date
6	8750	1-8-2000	8-1-2000	2000-1-8-12-49-10.9784529209137	2000-01-08	Date
7	7736	1-9-2000	9-1-2000	2000-1-9-12-49-10.9784529209137	2000-01-09	Date
1	11418	1-10-2000	10-1-2000	2000-1-10-12-49-10.9784529209137	2000-01-10	Date

mdy(): parses in order- month, date, year. I will apply to the “mdy.raw” column and save the results as “mdy.clean.” The mdy() function will save the results as year-month-date. This is helpful when trying to normalize data. As you can see below, the “ymd.clean” =“mdy.clean”

cleaned_date_df <- cleaned_date_df%>%
                  mutate(mdy.clean = mdy(mdy.raw))%>%
                  select(-c(mdy.raw))
kable(head(cleaned_date_df,10),  caption = "mdy() results")

mdy() results
day_of_week	births	dmy.raw	ymd.hms.raw	ymd.clean	ymd.type	mdy.clean
6	8843	1-1-2000	2000-1-1-12-49-10.9784529209137	2000-01-01	Date	2000-01-01
7	7816	2-1-2000	2000-1-2-12-49-10.9784529209137	2000-01-02	Date	2000-01-02
1	11123	3-1-2000	2000-1-3-12-49-10.9784529209137	2000-01-03	Date	2000-01-03
2	12703	4-1-2000	2000-1-4-12-49-10.9784529209137	2000-01-04	Date	2000-01-04
3	12240	5-1-2000	2000-1-5-12-49-10.9784529209137	2000-01-05	Date	2000-01-05
4	12260	6-1-2000	2000-1-6-12-49-10.9784529209137	2000-01-06	Date	2000-01-06
5	12280	7-1-2000	2000-1-7-12-49-10.9784529209137	2000-01-07	Date	2000-01-07
6	8750	8-1-2000	2000-1-8-12-49-10.9784529209137	2000-01-08	Date	2000-01-08
7	7736	9-1-2000	2000-1-9-12-49-10.9784529209137	2000-01-09	Date	2000-01-09
1	11418	10-1-2000	2000-1-10-12-49-10.9784529209137	2000-01-10	Date	2000-01-10

dmy(): parses in order- day, month, year. The dmy() function will save the results as year-month-date. This is helpful when trying to normalize data. As you can see below, the “ymd.clean” = “dmy.clean”

cleaned_date_df <- cleaned_date_df%>%
                  mutate(dmy.clean = dmy(dmy.raw))%>%
                  select(-c(dmy.raw))
kable(head(cleaned_date_df,10),  caption = "dmy() results")

dmy() results
day_of_week	births	ymd.hms.raw	ymd.clean	ymd.type	mdy.clean	dmy.clean
6	8843	2000-1-1-12-49-10.9784529209137	2000-01-01	Date	2000-01-01	2000-01-01
7	7816	2000-1-2-12-49-10.9784529209137	2000-01-02	Date	2000-01-02	2000-01-02
1	11123	2000-1-3-12-49-10.9784529209137	2000-01-03	Date	2000-01-03	2000-01-03
2	12703	2000-1-4-12-49-10.9784529209137	2000-01-04	Date	2000-01-04	2000-01-04
3	12240	2000-1-5-12-49-10.9784529209137	2000-01-05	Date	2000-01-05	2000-01-05
4	12260	2000-1-6-12-49-10.9784529209137	2000-01-06	Date	2000-01-06	2000-01-06
5	12280	2000-1-7-12-49-10.9784529209137	2000-01-07	Date	2000-01-07	2000-01-07
6	8750	2000-1-8-12-49-10.9784529209137	2000-01-08	Date	2000-01-08	2000-01-08
7	7736	2000-1-9-12-49-10.9784529209137	2000-01-09	Date	2000-01-09	2000-01-09
1	11418	2000-1-10-12-49-10.9784529209137	2000-01-10	Date	2000-01-10	2000-01-10

ymd_hms(): parses in order- year, month, day _ hour, minute, second

cleaned_date_df <- cleaned_date_df%>%
                  mutate(ymd.hms.clean = ymd_hms(ymd.hms.raw))%>%
                  select(-c(ymd.hms.raw))
kable(head(cleaned_date_df,10),  caption = "ymd_hms() results")

ymd_hms() results
day_of_week	births	ymd.clean	ymd.type	mdy.clean	dmy.clean	ymd.hms.clean
6	8843	2000-01-01	Date	2000-01-01	2000-01-01	2000-01-01 12:49:10
7	7816	2000-01-02	Date	2000-01-02	2000-01-02	2000-01-02 12:49:10
1	11123	2000-01-03	Date	2000-01-03	2000-01-03	2000-01-03 12:49:10
2	12703	2000-01-04	Date	2000-01-04	2000-01-04	2000-01-04 12:49:10
3	12240	2000-01-05	Date	2000-01-05	2000-01-05	2000-01-05 12:49:10
4	12260	2000-01-06	Date	2000-01-06	2000-01-06	2000-01-06 12:49:10
5	12280	2000-01-07	Date	2000-01-07	2000-01-07	2000-01-07 12:49:10
6	8750	2000-01-08	Date	2000-01-08	2000-01-08	2000-01-08 12:49:10
7	7736	2000-01-09	Date	2000-01-09	2000-01-09	2000-01-09 12:49:10
1	11418	2000-01-10	Date	2000-01-10	2000-01-10	2000-01-10 12:49:10

cleaned_date_df <- cleaned_date_df%>%
                  mutate(wday.clean = wday(ymd.clean))%>%
                  select(-c(day_of_week))
kable(head(cleaned_date_df,10),  caption = "wday() results")

wday() results
births	ymd.clean	ymd.type	mdy.clean	dmy.clean	ymd.hms.clean	wday.clean
8843	2000-01-01	Date	2000-01-01	2000-01-01	2000-01-01 12:49:10	7
7816	2000-01-02	Date	2000-01-02	2000-01-02	2000-01-02 12:49:10	1
11123	2000-01-03	Date	2000-01-03	2000-01-03	2000-01-03 12:49:10	2
12703	2000-01-04	Date	2000-01-04	2000-01-04	2000-01-04 12:49:10	3
12240	2000-01-05	Date	2000-01-05	2000-01-05	2000-01-05 12:49:10	4
12260	2000-01-06	Date	2000-01-06	2000-01-06	2000-01-06 12:49:10	5
12280	2000-01-07	Date	2000-01-07	2000-01-07	2000-01-07 12:49:10	6
8750	2000-01-08	Date	2000-01-08	2000-01-08	2000-01-08 12:49:10	7
7736	2000-01-09	Date	2000-01-09	2000-01-09	2000-01-09 12:49:10	1
11418	2000-01-10	Date	2000-01-10	2000-01-10	2000-01-10 12:49:10	2
###Additi	onal Lubridat	e Functions	:

Extract Information From Dates:

tody() : returns today’s date
year(): returns the year
month(): returns the month

Sys.getlocale()

## [1] "LC_COLLATE=English_United States.1252;LC_CTYPE=English_United States.1252;LC_MONETARY=English_United States.1252;LC_NUMERIC=C;LC_TIME=English_United States.1252"

this_day <- today()
this_day

## [1] "2018-05-09"

Date-Time Information:

these were used to create the base dataframe

day(): returns the day
wday(): day of the week
now(): date-time of the exact moment
hour(): hour of the exact moment
minute(): minute of the exact moment
second(): second of the exact moment