permute lm and output the results as a tidy table.
tidy_lm_bootstrap_permute(data, formula, n_bootstrap = 1000, bootstrap_ci = 0.95, n_permute = 1000, var_permute, r_squared_ci = 0.95)
| data | a data.frame to be analyzed |
|---|---|
| formula | a formula to be analyzed as typically written for the |
| n_bootstrap | = 1000 (default) the number of bootstraps to perform |
| bootstrap_ci | = 0.95 (default) |
| n_permute | = 1000 (default) the number of permutations to perform |
| var_permute | variable(s) to unlink in the permutation |
| r_squared_ci | = 0.95 (default) |
outputs tidy table that includes the p.value from the permutation of a lm test
packages <- c("broom", "broomExtra", "dplyr", "modelr", "purrr", "tibble") xfun::pkg_attach2(packages, message = F) data <- tibble( a = scale(sample.int(100), scale = F), b = scale(sample.int(100), scale = F), c = b^2, d = scale(sample.int(100), scale = F) ) tidy_lm_bootstrap_permute(data, a ~ b, var_permute = "a")#> Source: local data frame [2 x 14] #> Groups: <by row> #> #> # A tibble: 2 x 14 #> term estimate std.error statistic p.value estimate.ci.0.9… estimate.ci.0.9… #> <chr> <dbl> <dbl> <dbl> <dbl> <dbl> <dbl> #> 1 (Int… 1.14e-16 2.89 3.96e-17 1 -5.79 5.62 #> 2 b 1.43e- 1 0.1000 1.43e+ 0 0.156 -0.0396 0.335 #> # … with 7 more variables: p.permute.1000 <dbl>, #> # r.squared.ci.0.95.ll.boot.1000 <dbl>, r.squared.ci.0.95.ul.boot.1000 <dbl>, #> # adj.r.squared.ci.0.95.ll.boot.1000 <dbl>, #> # adj.r.squared.ci.0.95.ul.boot.1000 <dbl>, r.squared <dbl>, #> # adj.r.squared <dbl>