Influence plots: Two diagnostic plots for checking p-value influencers

Description

Two different plot types that visualize p-value influencers.

1. inflPlot: plots the linear regression, marks the reverser(s) in darkred and displays trend lines for the full and leave-reversers-out data set (black and darkred, respectively).
2. pvalPlot: plots the p-values for each leave-one-out data point and displays the (log) p-values as an index plot with reverser points in darkred, together with the \(\alpha\)-border as defined in lmInfl and the original models' p-value.

Usage

inflPlot(infl, measure, ...) 
pvalPlot(infl, ...)

Value

The corresponding plot.

Arguments

infl: an object obtained from lmInfl.
measure: which influence measure to use, see 'Details'.
...: other plotting parameters.

Author

Andrej-Nikolai Spiess

Details

The influence measure's to use are those listed in lmInfl, with the following syntax:
"dfb.Slope", "dffit", "cov.r", "cook.d", "hat", "sR", "hadi", "cdr", "Si".

References

Regression diagnostics: Identifying influential data and sources of collinearity.
Belsley DA, Kuh E, Welsch RE.
John Wiley, New York (2004).

Applied Regression Analysis: A Research Tool.
Rawlings JO, Pantula SG, Dickey DA.
Springer; 2nd Corrected ed. 1998. Corr. 2nd printing 2001.

Applied Regression Analysis and Generalized Linear Models.
Fox J.
SAGE Publishing, 3rd ed, 2016.

Residuals and Influence in Regression.
Cook RD & Weisberg S.
Chapman & Hall, 1st ed, New York, USA (1982).

Examples

Run this code

set.seed(123)
a <- 1:20
b <- 5 + 0.08 * a + rnorm(20, 0, 1)
LM1 <- lm(b ~ a)
res1 <- lmInfl(LM1) 
inflPlot(res1)
pvalPlot(res1)

Run the code above in your browser using DataLab