lincom | R Documentation |
Produces point estimates, interval estimates, and p-values for linear
combinations of regression coefficients using a uRegress
object.
lincom( reg, comb, null.hypoth = 0, conf.level = 0.95, robustSE = TRUE, joint.test = FALSE, useFdstn = FALSE, eform = reg$fnctl != "mean" )
reg |
an object of class |
comb |
a vector or matrix containing the values of the constants which create the linear combination of the form c_0 + c_1β_1 + … Zeroes must be given if coefficients aren't going to be included. For testing multiple combinations, this must be a matrix with number of columns equal to the number of coefficients in the model. |
null.hypoth |
the null hypothesis to compare the linear combination of
coefficients against. This is a scalar if one combination is given, and a
vector or matrix otherwise. The default value is |
conf.level |
a number between 0 and 1, indicating the desired confidence level for intervals. |
robustSE |
a logical value indicating whether or not to use robust
standard errors in calculation. Defaults to |
joint.test |
a logical value indicating whether or not to use a joint Chi-square test
for all the null hypotheses. If joint.test is |
useFdstn |
a logical indicator that the F distribution should be used for test statistics
instead of the chi squared distribution. Defaults to |
eform |
a logical value indicating whether or not to exponentiate the estimated coefficient. By default this is performed based on the type of regression used. |
A list of class lincom
(joint.test
is False
) or
lincom.joint
(joint.test
is True
). For the lincom
class,
comb
entries in the list are labeled comb1
, comb2
, etc. for as many linear combinations were used.
Each is a list with the following components:
printMat |
A formatted table with inferential results for the linear combination of coefficients. These include the point estimate, standard error, confidence interval, and t-test for the linear combination. |
nms |
The name of the linear combination, for printing. |
null.hypoth |
The null hypothesis for the linear combination. |
# Loading required libraries library(sandwich) # Reading in a dataset data(mri) # Linear regression of LDL on age (with robust SE by default) testReg <- regress ("mean", ldl~age+stroke, data = mri) # Testing coefficient created by .5*age - stroke (the first 0 comes from excluding the intercept) testC <- c(0, 0.5, -1) lincom(testReg, testC) # Test multiple combinations: # whether separately whether .5*age - stroke = 0 or Intercept + 60*age = 125 testC <- matrix(c(0, 0.5, -1, 1, 60, 0), byrow = TRUE, nrow = 2) lincom(testReg, testC, null.hypoth = c(0, 125)) # Test joint null hypothesis: # H0: .5*age - stroke = 0 AND Intercept + 60*age = 125 lincom(testReg, testC, null.hypoth = c(0, 125), joint.test = TRUE)
Add the following code to your website.
For more information on customizing the embed code, read Embedding Snippets.