# incrmpe
> Compute the [mean percentage error][mean-percentage-error] (MPE) incrementally.
The [mean percentage error][mean-percentage-error] is defined as
where `f_i` is the forecast value and `a_i` is the actual value.
## Usage
```javascript
var incrmpe = require( '@stdlib/stats/incr/mpe' );
```
#### incrmpe()
Returns an accumulator `function` which incrementally computes the [mean percentage error][mean-percentage-error].
```javascript
var accumulator = incrmpe();
```
#### accumulator( \[f, a] )
If provided input values `f` and `a`, the accumulator function returns an updated [mean percentage error][mean-percentage-error]. If not provided input values `f` and `a`, the accumulator function returns the current [mean percentage error][mean-percentage-error].
```javascript
var accumulator = incrmpe();
var m = accumulator( 2.0, 3.0 );
// returns ~33.33
m = accumulator( 1.0, 4.0 );
// returns ~54.17
m = accumulator( 3.0, 5.0 );
// returns ~49.44
m = accumulator();
// returns ~49.44
```
## Notes
- Input values are **not** type checked. If provided `NaN` or a value which, when used in computations, results in `NaN`, the accumulated value is `NaN` for **all** future invocations. If non-numeric inputs are possible, you are advised to type check and handle accordingly **before** passing the value to the accumulator function.
- Be careful when interpreting the [mean percentage error][mean-percentage-error] as errors can cancel. This stated, that errors can cancel makes the [mean percentage error][mean-percentage-error] suitable for measuring the bias in forecasts.
- **Warning**: the [mean percentage error][mean-percentage-error] is **not** suitable for intermittent demand patterns (i.e., when `a_i` is `0`). Interpretation is most straightforward when actual and forecast values are positive valued (e.g., number of widgets sold).
## Examples
```javascript
var randu = require( '@stdlib/random/base/randu' );
var incrmpe = require( '@stdlib/stats/incr/mpe' );
var accumulator;
var v1;
var v2;
var i;
// Initialize an accumulator:
accumulator = incrmpe();
// For each simulated datum, update the mean percentage error...
for ( i = 0; i < 100; i++ ) {
v1 = ( randu()*100.0 ) + 50.0;
v2 = ( randu()*100.0 ) + 50.0;
accumulator( v1, v2 );
}
console.log( accumulator() );
```
[mean-percentage-error]: https://en.wikipedia.org/wiki/Mean_percentage_error
