---
title: "stratRandSample"
date: "`r Sys.Date()`"
output: rmarkdown::html_vignette
vignette: >
  %\VignetteIndexEntry{stratRandSample}
  %\VignetteEngine{knitr::rmarkdown}
  %\VignetteEncoding{UTF-8}
---

```{r setup, include = FALSE}
knitr::opts_chunk$set(
  collapse = TRUE,
  comment = "#>"
)
library(coreStatsNMR)
library(dplyr)
library(data.table)

set.seed(1234)
```

`stratRandSample` is a function for generating a random stratified sample from a dataset. It takes up to 6 input values. The first three are required: a dataframe (or data.table), a variable listing strata, another specifying the number (or proportion) of units per stratum. The remaining three are optional: `select` allows for subsetting strata in line with the sampling, replace toggles sampling with replacement, and `bothSets` toggles whether the function returns both the sampled and non-sampled portions of the dataset.

### Sample data

For testing, we use the `mtcars` dataset that comes stock with R. It contains data from the 1974 Motor Trend US magazine, qith fuel economy and other statistics for a set of 32 cars. It looks something like this:

```{r mtcars}
glimpse(mtcars)
```

We add random weights to the data for testing below, and force a few columns into factors for the sake of our examples.

```{r test_data, echo=TRUE}
set.seed(1234)

rand_wt <- data.frame(rand_bin = 1:3,
                      rand_wt = rnorm(3, mean = 1, sd = 0.333))

test_data <- mtcars %>% 
  mutate(index = row_number(),
         rand_bin = sample(1:3, nrow(mtcars), replace = TRUE)) %>% 
  merge(rand_wt,
        by = "rand_bin") %>% 
  mutate(across(c(cyl, am, vs, gear, carb), as.factor))

test_dt <- data.table(test_data)
```

The data has `r nrow(test_data)` rows.

```{r}
nrow(test_data)
```

#### Random 80% of dataset

```{r simple_rand}
set.seed(1234)
stratRandSample(test_data, size = 0.8) %>% 
  nrow()

stratRandSample(test_dt, size = 0.8) %>% 
  nrow()
```


#### Random 30% of each group (# of cylinders)

```{r strat_sampleA}
stratRandSample(test_data, "cyl", 0.3) %>% 
  count(cyl)
stratRandSample(test_dt, "cyl", 0.3) %>% 
  count(cyl)
```

#### Random sample, different % per stratum

- 50% of 4-cylinder cars
- 30% of 6-cylinder cars
- 10% of 8-cylinder cars

```{r strat_sampleB}
stratRandSample(test_data, "cyl",
                size = c("4" = 0.5, "6" = 0.3, "8" = 0.1)) %>% 
  count(cyl)
stratRandSample(test_dt, "cyl",
                size = c("4" = 0.5, "6" = 0.3, "8" = 0.1)) %>% 
  count(cyl)
```

#### Misc. things

Automatically rounds to integer number of rows (e.g. 0.081*11 produces 1 row)
```{r truncA}
stratRandSample(test_data, "cyl",
                size = c("4" = 0.081, "6" = 0.407, "8" = 0.513)) %>% 
  count(cyl)
```

```{r truncB}
stratRandSample(test_dt, "cyl",
                size = c("4" = 0.09, "6" = 0.49, "8" = 0.59)) %>% 
  count(cyl)
```