Fast linear regression by group

Question

I have 500K users and I need to compute a linear regression (with intercept) for each of them.

Each user has around 30 records.

I t

Answer 1

An example using Rfast.

Assuming a single response and 500K predictor variables.

y <- rnorm(30)
x <- matrnorm(500*1000,30)
system.time( Rfast::univglms(y, x,"normal") )  ## 0.70 seconds

Assuming 500K response variables and a singl predictor variable.

system.time( Rfast::mvbetas(x,y) )  ## 0.60 seconds

Note: The above times will decrease in the nearby future.