Bootstrapping the data
Using the infer package with type = "bootstrap", you can repeatedly sample from the dataset to estimate the sampling distribution and standard error of the slope coefficient. Using the sampling distribution will allow you to directly find a confidence interval for the underlying population slope.
Bu egzersiz
Inference for Linear Regression in R
kursunun bir parçasıdırEgzersiz talimatları
Use the infer steps to bootstrap the twins data 1000 times. You don't have to hypothesize() because now you're creating confidence intervals, not hypothesis testing!
- Specify
FosterversusBiological. - Generate
1000replicates by bootstrapping. - Calculate the slope statistic.
Uygulamalı interaktif egzersiz
Bu örnek kodu tamamlayarak bu egzersizi bitirin.
# Set the seed for reproducibility
set.seed(4747)
# Calculate 1000 bootstrapped slopes
boot_slope <- twins %>%
# Specify Foster vs. Biological
___ %>%
# Generate 1000 bootstrap replicates
___ %>%
# Calculate the slope statistic
___
# See the result
boot_slope