Regression with time and date

Hey guys,

as a beginner I need some help with my task.

I need to explain from my marketing data how important the type of TV spot is in relation to placement and time of broadcast.

I already converted the date for example:

2020-12-25 17:31:37

to Weekdays (monday, tuesday etc.)

But now I have difficulties with making a regression with the Spots 1-5 (as a dependent variable) and time and weekdays as a independent variable. How do I "extract" time out of the example above? It always say that Variable lengths are different

Hope you get the point.


Probably using lubridate package which has convenience functions for extracting hours/seconds/minutes from a time vector would help you.
Do more with dates and times in R (

A Reprex from you would be useful in order to provide you specific help
FAQ: How to do a minimal reproducible example ( reprex ) for beginners

1 Like

you don't need dummy variables, lm would make them automatically.

#made up data
(wd <- weekdays(seq(from=Sys.Date()-99,to=Sys.Date(),by=1)))
(someval <- 20*nchar(wd) +rnorm(100,1,1))

#look at it

#do regression
(lm1 <- lm(someval ~ wd))
1 Like

This topic was automatically closed 21 days after the last reply. New replies are no longer allowed.

If you have a query related to it or one of the replies, start a new topic and refer back with a link.