OLSPlots.jl

OLSPlots.jl provides diagnostic plots in Julia for ordinary least squares  
regression models prepared with GLM.lm. It's based on the plots method for the R  
Programming Language's lm regression models. It differs slightly in the rendering  
of loess lines due to a different algorithm in Julia's Loess package compared to  
the R equivalent.

Compared to R

Using a simplified mtcars data set from R available here, the R implementation is as follows:

model = lm(mpg ~ hp + drat + wt, df)  
plot(model)

The corresponding Julia implementation is as follows:

using CSV, DataFrames, GLM, OLSPlots  
mtcars = ("data/mtcars.csv", DataFrames)  
model = lm(@formula(mpg ~ hp + drat + wt), mtcars)  
diagnostic_plots(model)

diagnostic_plots takes a which argument, a vector in the range 1:6 to selectively display plots. The default is [1,2,3,5]. To show the additional plots

diagnostic_plots(model, which = [4,6])

Additional details are in the documentation.