Extracting portions of one vector to create others

This is an extremely common transformation in regression, and it has a dedicated function (model.matrix) for accomplishing it in R. It does presume you have a bit of familiarity with the formula literal type in R... which is created using the tilde character (e.g. ~ CLASS). Also, since including an intercept term in a regression is the default, if you only want the columns you described then you need to inform the function that you want to leave the intercept out (~ CLASS - 1). It doesn't apply this discrete transformation unless the referenced variable is a factor variable, so I convert it from integer type in the short example below.

Beth <- data.frame(

??CLASS = c(
????1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 2, 2, 2, 2, 2
????, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 4, 4, 4, 4, 4, 4, 4, 4
????, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7
????, 7, 7, 7, 7, 7, 7, 7, 7, 9, 9, 9, 9, 9, 9, 9, 9, 9, 9, 9, 9, 9, 9, 1))
Beth$CLASS <- as.character(Beth$CLASS)
result <- model.matrix( ~ CLASS - 1, Beth)
result

Note that the result is a matrix, not a data frame so you will have to use brackets and a blank for the row spec if you want to access one column at a time: 

result[, "CLASS1"]

Refer to the help page for more on this function:
?model.matrix

Extracting portions of one vector to create others

Thread (8 messages)