Skip to contents

Asymptotic relative efficiency of two estimators

Source: R/areN.R
Areratio.Rd

ARE(T2, T1) = (var1/n1) / (var2/n2); the normal benchmarks 2/pi (median vs mean) and 3/pi (Hodges-Lehmann vs mean) are returned as closed forms.

Usage

Areratio(var1, var2, n1 = 1, n2 = 1)

Arguments

var1, var2

Asymptotic variances, strictly positive.

n1, n2

Sample sizes the variances refer to.

Value

List with are, logare, var1, var2, normalmedian, normalhl.

References

Hodges and Lehmann (1956), Annals of Mathematical Statistics 27(2), 324-335; standard published form.

Examples

Areratio(var1 = 5L, var2 = 5L)
#> $are
#> [1] 1
#> 
#> $logare
#> [1] 0
#> 
#> $var1
#> [1] 5
#> 
#> $var2
#> [1] 5
#> 
#> $normalmedian
#> [1] 0.6366198
#> 
#> $normalhl
#> [1] 0.9549297
#> 
#> $method
#> [1] "Asymptotic relative efficiency (Hodges-Lehmann 1956)"
#> 
#> attr(,"class")
#> [1] "morie_rich_result" "list"