Using Rmarkdown

```{r} library(tidyverse) ```

Consider the function `slr` which returns the intercept and slope of a simple linear regression model.

```{r} # get a column of data extract <- function(data, x) { pull(select(data, x)) } ```

```{r} mean_narm <- function(data, y) { mean(extract(data, y), na.rm = TRUE) } ```

```{r} dot <- function(y, x) { sum(x * y) } ```

```{r} slr <- function(data) { x <- data$x y <- data$y mux <- mean_narm(data, "x") muy <- mean_narm(data, "y") sxy <- dot(x - mux, y - muy) sxx <- dot(x - mux, x - mux) slope <- sxy / sxx intercept <- muy - slope * mux c(intercept, slope) } ```

Despite of the inefficiency of the code, this code is erroneous and doesn't return the correct result.

```{r} df <- tibble(x = rnorm(10), y = rnorm(10)) all.equal( slr(df), coef(lm(y ~ x, data = df)), check.attributes = FALSE) ```

##### Could you find the culprit (explain it) and provide a targeted fix (minimal changes to the code)?

Hint: you could use `browser()`, `debug`/`undebug` or `debugonce` to debug the functions.

```{r} stopifnot( all.equal( slr(df), coef(lm(y ~ x, data = df)), check.attributes = FALSE) ) ```