Learn R Programming

lsr (version 0.5.2)

standardCoefs: Standardised regression coefficients

Description

Calculates the standardised regression coefficients for a linear model.

Usage

standardCoefs(x)

Arguments

x

A linear model object (i.e. class lm)

Value

A matrix with the regressors as rows, and the two different regression coefficients (unstandardised and standardised) as the two columns. The columns are labeled b (unstandardised) and beta (standardised).

Details

Calculates the standardised regression coefficients (beta-weights), namely the values of the regression coefficients that would have been observed has all regressors and the outcome variable been scaled to have mean 0 and variance 1 before fitting the regression model. Standardised coefficients are sometimes useful in some applied contexts since there is a sense in which all beta values are "on the same scale", though this is not entirely unproblematic.

Examples

Run this code
# NOT RUN {
# Example 1: simple linear regression

# data
X1 <- c(0.69, 0.77, 0.92, 1.72, 1.79, 2.37, 2.64, 2.69, 2.84, 3.41)
Y  <- c(3.28, 4.23, 3.34, 3.73, 5.33, 6.02, 5.16, 6.49, 6.49, 6.05)

model1 <- lm( Y ~ X1 )  # run a simple linear regression
coefficients( model1 )  # extract the raw regression coefficients
standardCoefs( model1 ) # extract standardised coefficients


# Example 2: multiple linear regression

X2 <- c(0.19, 0.22, 0.95, 0.43, 0.51, 0.04, 0.12, 0.44, 0.38, 0.33)
model2 <- lm( Y ~ X1 + X2 )   # new model
standardCoefs( model2 )       # standardised coefficients

#Example 3: interaction terms

model3 <- lm( Y ~ X1 * X2 )
coefficients( model3 )
standardCoefs( model3 )

# Note that these beta values are equivalent to standardising all
# three regressors including the interaction term X1:X2, not merely
# standardising the two predictors X1 and X2.

# }

Run the code above in your browser using DataLab