In MATHSTATSOU/Intro2R: Introduction to R

knitr::opts_chunk$set(
  collapse = TRUE,
  comment = "#>"
)

library(Intro2R)

Introduction

Suppose we have a paired data set then we can test the NULL hypothesis

$$ H_0 : \mu_1 - \mu_2 = 0 $$

Using t.test(x,y,mu = 0, paired = TRUE)

Note that the order of the data matters, in the above t.test mu=0 means $\mu_x - \mu_y=0$ whereas t.test(y,x,mu = 0, paired = TRUE), mu=0 means $\mu_y -\mu_x = 0$.

The NULL hypothesis could be expressed differently depending on the experiment and prior information eg: $\mu_x - \mu_y = 10$ rather than 0.

Where does a paired sample come from?

Paired data are very common. Whenever multivariate data is measured pairs will be formed.

MATHSTATSOU/Intro2R documentation built on Feb. 20, 2025, 6:18 a.m.