-1

I calculated a linear model, with the lm() function. I know how to extract the residuals, using the resid() function, but I would like to calculate the values of the residuals variance by year and plot it.

Mateus Maciel
  • 151
  • 1
  • 1
  • 10
  • 1
    Please provide a [minimal reproducible example](https://stackoverflow.com/questions/5963269/how-to-make-a-great-r-reproducible-example) and show us the steps you've already taken – starja Jul 24 '20 at 21:31

1 Answers1

1

You could use a data frame out of the residuals and the year, aggreagte it and make a barplot.

fdf <- data.frame(resid=fit$residuals, year=fit$model$year)
res <- aggregate(resid ~ year, fdf, var)
barplot(resid ~ year, res)

enter image description here

Toy data

set.seed(42)
dat <- transform(cbind.data.frame(year=1:5, x=rnorm(500)), y=.5 * x + rnorm(500))
fit <- lm(y ~ x + year, dat)
jay.sf
  • 60,139
  • 8
  • 53
  • 110