06-functions.Rmd

## Functions

In math, we use functions all the time. Functions have inputs and outputs, e.g. $f(x) = 3x^2+x+1$

R can be considered to be a functional language. These are functions that we have just used, `rnorm(), mean(), sd()`. 

What does a function in R look like? The polynomial function above would look like this in R:

```
mypolyfn <- function(x) {
  f <- 3*x^2 + x + 1
  return(f)
}
```

It takes `x` as the input, and returns `f` as output. The benefit of functions is that it makes your work more efficient - anything that you need to do again and again, or on different sets of data, should be programmed into a function.

*Try it!* Write the function below, so that it is defined. And then test it out by inputting different values of `x`, e.g. `mypolyfn(2)`, or `mypolyfn(-1)`, or even entering a vector of `x`'s, `mypolyfn(seq(-5, 5, 1))`

```{r functions, exercise=TRUE}

```

What can go wrong?

This function is expecting `x` to be a number. If the user inputs something other that a number, what happens? 

*Try it!* Try running your function with a text string of your team name instead of a number, e.g. `mypolyfn("numbat")`.

You get an error. Well-written functions check for silly user errors. A better version of the function would be:

```
mypolyfn <- function(x) {
  stopifnot(is.numeric(x))
  f <- 3*x^2 + x + 1
  return(f)
}
```