I have data that looks like this :
head(df) player monday tuesday wednesday thursday friday saturday <chr> <dbl> <dbl> <dbl> <dbl> <dbl> <dbl> 1 Jack 15 12 NA 6 NA NA 2 John 22 18 7 NA NA NA 3 Peter NA NA NA NA 17.4 29.7 4 Jan NA 10 NA 4 NA NA 5 Mike 32 NA NA 18.7 NA NA 6 Jason NA NA NA NA 11 14
Every player performs at a different level. What I want to know is the effect that day has on performance as increase/decrease proportion of a base. So if we assume base Monday for Jack, Tuesday is 0.8 of Monday's performance.
My thought is that I need to fit coefficients for each day using least squares of the difference between 1 and the ratio of pairwise performance numbers. But I'm not sure how to do that.
Although maybe I'm over complicating, and there's a better way of tackling this.
Any help would be much appreciated.