Ivan Jacob Agaloos Pesigan 2023-11-30
Generates robust confidence intervals for standardized regression coefficients using heteroskedasticity-consistent standard errors for models fitted by
lm() as described in Dudgeon (2017: http://doi.org/10.1007/s11336-017-9563-z). A description of the package and code examples are presented in Pesigan, Sun, and Cheung (2023: https://doi.org/10.1080/00273171.2023.2201277).
You can install the CRAN release of
You can install the development version of
betaSandwich from GitHub with:
In this example, a multiple regression model is fitted using program quality ratings (
QUALITY) as the regressand/outcome variable and number of published articles attributed to the program faculty members (
NARTIC), percent of faculty members holding research grants (
PCTGRT), and percentage of program graduates who received support (
PCTSUPP) as regressor/predictor variables using a data set from 1982 ratings of 46 doctoral programs in psychology in the USA (National Research Council, 1982). Robust confidence intervals for the standardized regression coefficients are generated using the
BetaHC() function from the
betaSandwich package following Dudgeon (2017).
df <- betaSandwich::nas1982
object <- lm(QUALITY ~ NARTIC + PCTGRT + PCTSUPP, data = df)
Estimate the standardized regression slopes and the corresponding robust sampling covariance matrix.
BetaHC(object, type = "hc3", alpha = 0.05) #> Call: #> BetaHC(object = object, type = "hc3", alpha = 0.05) #> #> Standardized regression slopes with HC3 standard errors: #> est se t df p 2.5% 97.5% #> NARTIC 0.4951 0.0786 6.3025 42 0.0000 0.3366 0.6537 #> PCTGRT 0.3915 0.0818 4.7831 42 0.0000 0.2263 0.5567 #> PCTSUPP 0.2632 0.0855 3.0786 42 0.0037 0.0907 0.4358
The package can also be used to generate confidence intervals for R-squared, adjusted R-squared, and differences of standardized regression coefficients.
See GitHub Pages for package documentation.