Joint location-and-scale test using Fisher's method

jlsp performs the joint location-and-scale test using Fisher's method.

Usage

jlsp(y, x, covar = NULL, covar.var = FALSE, var.type = 1, x.sq = FALSE)

Arguments

y: vector of outcome values
x: vector of exposure values
covar: data.frame of covariates
covar.var: adjust the second stage (variance component) of the approach by the covariates
x.sq: include x-squared in the variance part of the model
type: type of test, where 1 = Breusch-Pagan variance test, and 2 = Brown-Forsythe variance test (default: 1)

Value

jlsp returns a list of results:

Q / F: the test statistic
DF: the degrees of freedom
P: the p-value

Author

James Staley jrstaley95@gmail.com

Examples

x <- rbinom(1000, 1, 0.5)
y <- 0.5 + 0.025 * x + rnorm(1000, 0, sqrt(0.005 * x)) + rnorm(1000, 0, 0.1)
jlsp(y, x, var.type = 2)
#> $location_test
#> $location_test$coef
#>              Estimate  Std. Error    t value     Pr(>|t|)
#> (Intercept) 0.5051985 0.004923175 102.616400 0.0000000000
#> x           0.0248402 0.007091233   3.502946 0.0004805401
#> 
#> $location_test$test
#>          F DF            P
#> 1 12.27063  1 0.0004805401
#> 
#> 
#> $scale_test
#> $scale_test$coef
#>               Estimate  Std. Error   t value      Pr(>|t|)
#> (Intercept) 0.08014668 0.002967853 27.004939 5.160697e-121
#> x           0.01835758 0.004274830  4.294343  1.923144e-05
#> 
#> $scale_test$test
#>          F DF            P
#> 1 18.44138  1 1.923144e-05
#> 
#> 
#> $location_scale_test
#>          Q DF            P
#> 1 36.99913  4 1.802048e-07
#>