Lastly I’d like to cover a very imprtant topic in data prep that is to handle date time data . Why the special attention? Because the way date time data is recorded is essentially as a character and when it is dealt with, its like numbers . Now parsing these seemingly character strings to numbers in a way that they can represent data time and then throw those time zones in the mix and you have pretty difficult situation to handle. Fret not , we have package lubridate to make our life pretty easy.

Please install package lubridate before you begin running these codes.

Note: All our examples are based on converting character strings to dates using functions. When you are reading data from a file, read a date time column as character and convert it to a date time object type later. Its always easier to do.

We’ll start with converting various dates in different formats stored as characters. These formats can be different in terms of with in that data in what order day, month and year appear.

Identify the order in which the year, month, and day appears in your dates. Now arrange “y”, “m”, and “d” in the same order. This is the name of the function in lubridate that will parse your dates. For example,

library(lubridate) 
ymd("20110604") 

## [1] "2011-06-04"

mdy("06-04-2011") 

## [1] "2011-06-04"

dmy("04/06/2011") 

## [1] "2011-06-04"

you include time components and timezones as well by simply adding the order of time components hours (“h”), minutes (“m”) and seconds (“s”). Appropriate function name exists.

arrive = ymd_hms("2011-06-04 12:00:00", tz = "Pacific/Auckland") 
leave = ymd_hms("2011-08-10 14:00:00", tz = "Pacific/Auckland")

you can extract and set individual elements of the date as well.

second(arrive)

So far we have seen seemingly well formatted character strings as input for dates which is not always the case . Dates can have months as character names or even abreviated form of three letter words. And same goes for other date components as well. You can handle that by specifying your own formats too these format builders and function parse_date_time.

• b : Abbreviated month name

• B : Full month name

• d : Day of the month as decimal number (01 to 31 or 0 to 31)

• H : Hours as decimal number (00 to 24 or 0 to 24). 24 hrs format

• I : Hours as decimal number (01 to 12 or 0 to 12). 12 hrs format

• j : Day of year as decimal number (001 to 366 or 1 to 366).

• m : Month as decimal number (01 to 12 or 1 to 12).

• M : Minute as decimal number (00 to 59 or 0 to 59).

• p : AM/PM indicator in the locale. Used in conjunction with I and not with H.

• S : Second as decimal number (00 to 61 or 0 to 61), allowing for up to two leap-seconds (but POSIXcompliant implementations will ignore leap seconds).

• OS :Fractional second.

• y : Year without century (00 to 99 or 0 to 99).

• Y : Year with century.

Although there are too many format builders here , you’ll generally use few. we’ll see example with those. you can handle hetrogenous formats as well.

parse_date_time("01-12-Jan","%d-%y-%b")

## [1] "2012-01-01 UTC"

parse_date_time("01-12-Jan 12:05 PM","%d-%y-%b %I:%M %p") 

## [1] "2012-01-01 12:05:00 UTC"