'How to do a simulation of a hypothesis test (using the Wald test) and report the Monte Carlo standard error of the power?
I am reading https://cran.r-project.org/web/packages/rsimsum/vignettes/A-introduction.html and trying to do a simulation of a hypothesis test (using the Wald test) and report the Monte Carlo standard error of the power. Assume that iid random sample X_i~ Bernoulli(\pi)
for i=1,\dots, 100
and sample size is 100
. We want to do a one side hypothesis test that
H_0: \pi\ge 0.75,\, H_a: \pi<0.75
.
Question: As in the above link, the power of a significant level alpha=0.05
is for a two side hypothesis test. How to use this function to do a one side hypothesis test? How to do such simulation using rsimsum
?
Also, when I do the example as in the link, it shows error 'there is no function simsum'
install.packages("rsimsum")
library(rsimsum)
data("MIsim", package = "rsimsum")
head(MIsim, n = 10)
str(MIsim)
s1 <- simsum(data = MIsim, estvarname = "b", true = 0.50, se = "se", methodvar = "method", ref = "CC")
Sources
This article follows the attribution requirements of Stack Overflow and is licensed under CC BY-SA 3.0.
Source: Stack Overflow
Solution | Source |
---|