DFFITS, the scaled deletion influence on the fitted value

Formula: DFFITS_i = t*_i sqrt(h_ii/(1-h_ii)), t*_i = e_i / (s_(i) sqrt(1-h_ii))

Usage

Dffitsols(X, y, intercept = TRUE)

Arguments

X: Design matrix.
y: Response.
intercept: Prepend a column of ones.

Value

List with dffits, cutoff, leverage, student, n, p.

References

Belsley, Kuh and Welsch (1980), Regression Diagnostics: Identifying Influential Data and Sources of Collinearity, Wiley. The definitions and the cutoffs used here (2/sqrt(n) for DFBETAS, 2 sqrt(p/n) for DFFITS, scaling by the delete-one root mean square s_(i)) are as documented by the SAS and R reference implementations, which cite BKW for them.

Examples

V <- c(1, 2, 3, 4, 5, 6, 7, 8)
Dffitsols(V, V)
#> $dffits
#> [1] -0.1847948  0.0000000  0.1733540  0.2920872  0.2920872  0.3620476  1.3375612
#> [8]  2.3445059
#> 
#> $cutoff
#> [1] 1
#> 
#> $leverage
#> [1] 0.4166667 0.2738095 0.1785714 0.1309524 0.1309524 0.1785714 0.2738095
#> [8] 0.4166667
#> 
#> $student
#> [1] -0.2186522  0.0000000  0.3718030  0.7524499  0.7524499  0.7765056  2.1782849
#> [8]  2.7740567
#> 
#> $n
#> [1] 8
#> 
#> $p
#> [1] 2
#> 
#> $method
#> [1] "DFFITS (Belsley-Kuh-Welsch)"
#> 
#> attr(,"class")
#> [1] "morie_rich_result" "list"