Estimate Cohen's f-Squared and Local f-Squared and the Corresponding Sampling Covariance Matrix

Estimate Cohen's f-Squared and Local f-Squared and the Corresponding Sampling Covariance Matrix

Usage

FSqDelta(object, alpha = c(0.05, 0.01, 0.001))

Arguments

object

Object of class betadelta, that is, the output of the BetaDelta() function.

alpha

Numeric vector. Significance level \(\alpha\).

Value

Returns an object of class esdelta which is a list with the following elements:

call

Function call.

fit

The argument object.

args

Function arguments.

vcov

Sampling covariance matrix of Cohen's f-squared (fsq) and the local f-squared of each regressor.

est

Vector of Cohen's f-squared (fsq) and the local f-squared of each regressor.

fun

Function used ("FSqDelta").

Details

The sampling covariance matrix is obtained by the delta method from the sampling covariance matrix of the variances and covariances stored in object (normal-theory or asymptotically distribution-free).

See also

Other Beta Delta Functions: BetaDelta(), DeltaRSqDelta(), DiffBetaDelta(), PCorDelta(), RSqDelta(), SCorDelta()

Author

Ivan Jacob Agaloos Pesigan

Examples

object <- lm(QUALITY ~ NARTIC + PCTGRT + PCTSUPP, data = nas1982)
std <- BetaDelta(object)
out <- FSqDelta(std)
# Methods -------------------------------------------------------
print(out)
#> Call:
#> FSqDelta(object = std)
#> 
#> Cohen's f-squared (fsq) and local f-squared with MVN standard errors:
#>            est     se      z      p   0.05%    0.5%    2.5%  97.5%  99.5%
#> fsq     4.1158 1.3531 3.0417 0.0024 -0.3367  0.6304  1.4637 6.7678 7.6012
#> NARTIC  0.9510 0.4017 2.3676 0.0179 -0.3707 -0.0836  0.1637 1.7382 1.9856
#> PCTGRT  0.6019 0.2896 2.0787 0.0376 -0.3509 -0.1439  0.0344 1.1694 1.3477
#> PCTSUPP 0.2909 0.1807 1.6099 0.1074 -0.3037 -0.1746 -0.0633 0.6452 0.7565
#>         99.95%
#> fsq     8.5682
#> NARTIC  2.2727
#> PCTGRT  1.5547
#> PCTSUPP 0.8856
summary(out)
#> Call:
#> FSqDelta(object = std)
#> 
#> Cohen's f-squared (fsq) and local f-squared with MVN standard errors:
#>            est     se      z      p   0.05%    0.5%    2.5%  97.5%  99.5%
#> fsq     4.1158 1.3531 3.0417 0.0024 -0.3367  0.6304  1.4637 6.7678 7.6012
#> NARTIC  0.9510 0.4017 2.3676 0.0179 -0.3707 -0.0836  0.1637 1.7382 1.9856
#> PCTGRT  0.6019 0.2896 2.0787 0.0376 -0.3509 -0.1439  0.0344 1.1694 1.3477
#> PCTSUPP 0.2909 0.1807 1.6099 0.1074 -0.3037 -0.1746 -0.0633 0.6452 0.7565
#>         99.95%
#> fsq     8.5682
#> NARTIC  2.2727
#> PCTGRT  1.5547
#> PCTSUPP 0.8856
coef(out)
#>       fsq    NARTIC    PCTGRT   PCTSUPP 
#> 4.1157764 0.9509843 0.6018924 0.2909455 
vcov(out)
#>                fsq      NARTIC        PCTGRT       PCTSUPP
#> fsq     1.83090366 0.292190379  0.1757962533  0.0810437672
#> NARTIC  0.29219038 0.161335262  0.0047781399  0.0012330834
#> PCTGRT  0.17579625 0.004778140  0.0838405969 -0.0003089663
#> PCTSUPP 0.08104377 0.001233083 -0.0003089663  0.0326604105
confint(out, level = 0.95)
#>               2.5 %    97.5 %
#> fsq      1.46373175 6.7678211
#> NARTIC   0.16373420 1.7382345
#> PCTGRT   0.03438011 1.1694047
#> PCTSUPP -0.06326298 0.6451539