In andrewpbray/infer: Tidy Statistical Inference

knitr::opts_chunk$set(fig.width = 8, fig.height = 5) 

Note: The type argument in generate() is automatically filled based on the entries for specify() and hypothesize(). It can be removed throughout the examples that follow. It is left in to reiterate the type of generation process being performed.

Data preparation

library(infer)
library(dplyr)
mtcars <- mtcars %>%
  mutate(cyl = factor(cyl),
         vs = factor(vs),
         am = factor(am),
         gear = factor(gear),
         carb = factor(carb))
# For reproducibility         
set.seed(2018)         

---

One numerical variable (mean)

mtcars %>%
  specify(response = mpg) %>% # formula alt: mpg ~ NULL
  hypothesize(null = "point", mu = 25) %>% 
  generate(reps = 100, type = "bootstrap") %>% 
  calculate(stat = "mean")

One numerical variable (median)

mtcars %>%
  specify(response = mpg) %>% # formula alt: mpg ~ NULL
  hypothesize(null = "point", med = 26) %>% 
  generate(reps = 100, type = "bootstrap") %>% 
  calculate(stat = "median")

One categorical (2 level) variable

mtcars %>%
  specify(response = am, success = "1") %>% # formula alt: am ~ NULL
  hypothesize(null = "point", p = .25) %>% 
  generate(reps = 100, type = "simulate") %>% 
  calculate(stat = "prop")

Two categorical (2 level) variables

mtcars %>%
  specify(am ~ vs, success = "1") %>% # alt: response = am, explanatory = vs
  hypothesize(null = "independence") %>%
  generate(reps = 100, type = "permute") %>%
  calculate(stat = "diff in props", order = c("0", "1"))

One categorical (>2 level) - GoF

mtcars %>%
  specify(cyl ~ NULL) %>% # alt: response = cyl
  hypothesize(null = "point", p = c("4" = .5, "6" = .25, "8" = .25)) %>%
  generate(reps = 100, type = "simulate") %>%
  calculate(stat = "Chisq")

Two categorical (>2 level) variables

mtcars %>%
  specify(cyl ~ am) %>% # alt: response = cyl, explanatory = am
  hypothesize(null = "independence") %>%
  generate(reps = 100, type = "permute") %>%
  calculate(stat = "Chisq")

One numerical variable one categorical (2 levels) (diff in means)

mtcars %>%
  specify(mpg ~ am) %>% # alt: response = mpg, explanatory = am
  hypothesize(null = "independence") %>%
  generate(reps = 100, type = "permute") %>%
  calculate(stat = "diff in means", order = c("0", "1"))

One numerical variable one categorical (2 levels) (diff in medians)

mtcars %>%
  specify(mpg ~ am) %>% # alt: response = mpg, explanatory = am
  hypothesize(null = "independence") %>%
  generate(reps = 100, type = "permute") %>%
  calculate(stat = "diff in medians", order = c("0", "1"))

One numerical one categorical (>2 levels) - ANOVA

mtcars %>%
  specify(mpg ~ cyl) %>% # alt: response = mpg, explanatory = cyl
  hypothesize(null = "independence") %>%
  generate(reps = 100, type = "permute") %>%
  calculate(stat = "F")

Two numerical vars - SLR

mtcars %>%
  specify(mpg ~ hp) %>% # alt: response = mpg, explanatory = cyl
  hypothesize(null = "independence") %>%
  generate(reps = 100, type = "permute") %>%
  calculate(stat = "slope")

One numerical variable (standard deviation)

Not currently implemented

mtcars %>%
  specify(response = mpg) %>% # formula alt: mpg ~ NULL
  hypothesize(null = "point", sigma = 5) %>% 
  generate(reps = 100, type = "bootstrap") %>% 
  calculate(stat = "sd")

Confidence intervals

One numerical (one mean)

mtcars %>%
  specify(response = mpg) %>%
  generate(reps = 100, type = "bootstrap") %>%
  calculate(stat = "mean")

One numerical (one median)

mtcars %>%
  specify(response = mpg) %>%
  generate(reps = 100, type = "bootstrap") %>%
  calculate(stat = "median")

One numerical (standard deviation)

mtcars %>%
  specify(response = mpg) %>%
  generate(reps = 100, type = "bootstrap") %>%
  calculate(stat = "sd")

One categorical (one proportion)

mtcars %>%
  specify(response = am, success = "1") %>%
  generate(reps = 100, type = "bootstrap") %>%
  calculate(stat = "prop")

One numerical variable one categorical (2 levels) (diff in means)

mtcars %>%
  specify(mpg ~ am) %>%
  generate(reps = 100, type = "bootstrap") %>%
  calculate(stat = "diff in means", order = c("0", "1"))

Two categorical variables (diff in proportions)

mtcars %>%
  specify(am ~ vs, success = "1") %>%
  generate(reps = 100, type = "bootstrap") %>%
  calculate(stat = "diff in props", order = c("0", "1"))

Two numerical vars - SLR

mtcars %>%
  specify(mpg ~ hp) %>% 
  generate(reps = 100, type = "bootstrap") %>%
  calculate(stat = "slope")

Two numerical vars - correlation

mtcars %>%
  specify(mpg ~ hp) %>% 
  generate(reps = 100, type = "bootstrap") %>%
  calculate(stat = "correlation")

andrewpbray/infer documentation built on Aug. 29, 2019, 5:57 a.m.