time-to-botec

Benchmark sampling in different programming languages
Log | Files | Refs | README

README.md (4248B)

      1 <!--
      2 
      3 @license Apache-2.0
      4 
      5 Copyright (c) 2018 The Stdlib Authors.
      6 
      7 Licensed under the Apache License, Version 2.0 (the "License");
      8 you may not use this file except in compliance with the License.
      9 You may obtain a copy of the License at
     10 
     11    http://www.apache.org/licenses/LICENSE-2.0
     12 
     13 Unless required by applicable law or agreed to in writing, software
     14 distributed under the License is distributed on an "AS IS" BASIS,
     15 WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
     16 See the License for the specific language governing permissions and
     17 limitations under the License.
     18 
     19 -->
     20 
     21 # incrmape
     22 
     23 > Compute the [mean absolute percentage error][mean-absolute-percentage-error] (MAPE) incrementally.
     24 
     25 <section class="intro">
     26 
     27 The [mean absolute percentage error][mean-absolute-percentage-error] is defined as
     28 
     29 <!-- <equation class="equation" label="eq:mean_absolute_percentage_error" align="center" raw="\operatorname{MAPE}  = \frac{100}{n} \sum_{i=0}^{n-1} \biggl| \frac{a_i - f_i}{a_i} \biggr|" alt="Equation for the mean absolute percentage error."> -->
     30 
     31 <div class="equation" align="center" data-raw-text="\operatorname{MAPE}  = \frac{100}{n} \sum_{i=0}^{n-1} \biggl| \frac{a_i - f_i}{a_i} \biggr|" data-equation="eq:mean_absolute_percentage_error">
     32     <img src="https://cdn.jsdelivr.net/gh/stdlib-js/stdlib@d4867162fd6445b10f93ca01f3f764bc646662d8/lib/node_modules/@stdlib/stats/incr/mape/docs/img/equation_mean_absolute_percentage_error.svg" alt="Equation for the mean absolute percentage error.">
     33     <br>
     34 </div>
     35 
     36 <!-- </equation> -->
     37 
     38 where `f_i` is the forecast value and `a_i` is the actual value.
     39 
     40 </section>
     41 
     42 <!-- /.intro -->
     43 
     44 <section class="usage">
     45 
     46 ## Usage
     47 
     48 ```javascript
     49 var incrmape = require( '@stdlib/stats/incr/mape' );
     50 ```
     51 
     52 #### incrmape()
     53 
     54 Returns an accumulator `function` which incrementally computes the [mean absolute percentage error][mean-absolute-percentage-error].
     55 
     56 ```javascript
     57 var accumulator = incrmape();
     58 ```
     59 
     60 #### accumulator( \[f, a] )
     61 
     62 If provided input values `f` and `a`, the accumulator function returns an updated [mean absolute percentage error][mean-absolute-percentage-error]. If not provided input values `f` and `a`, the accumulator function returns the current [mean absolute percentage error][mean-absolute-percentage-error].
     63 
     64 ```javascript
     65 var accumulator = incrmape();
     66 
     67 var m = accumulator( 2.0, 3.0 );
     68 // returns ~33.33
     69 
     70 m = accumulator( 1.0, 4.0 );
     71 // returns ~54.17
     72 
     73 m = accumulator( 3.0, 5.0 );
     74 // returns ~49.44
     75 
     76 m = accumulator();
     77 // returns ~49.44
     78 ```
     79 
     80 </section>
     81 
     82 <!-- /.usage -->
     83 
     84 <section class="notes">
     85 
     86 ## Notes
     87 
     88 -   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.
     89 
     90 -   **Warning**: the [mean absolute percentage error][mean-absolute-percentage-error]  has several shortcomings: 
     91 
     92     -   The measure is **not** suitable for intermittent demand patterns (i.e., when `a_i` is `0`).
     93     -   The [mean absolute percentage error][mean-absolute-percentage-error] is not symmetrical, as the measure cannot exceed 100% for forecasts which are too "low" and has no limit for forecasts which are too "high".
     94     -   When used to compare the accuracy of forecast models (e.g., predicting demand), the measure is biased toward forecasts which are too low. 
     95 
     96 </section>
     97 
     98 <!-- /.notes -->
     99 
    100 <section class="examples">
    101 
    102 ## Examples
    103 
    104 <!-- eslint no-undef: "error" -->
    105 
    106 ```javascript
    107 var randu = require( '@stdlib/random/base/randu' );
    108 var incrmape = require( '@stdlib/stats/incr/mape' );
    109 
    110 var accumulator;
    111 var v1;
    112 var v2;
    113 var i;
    114 
    115 // Initialize an accumulator:
    116 accumulator = incrmape();
    117 
    118 // For each simulated datum, update the mean absolute percentage error...
    119 for ( i = 0; i < 100; i++ ) {
    120     v1 = ( randu()*100.0 ) + 50.0;
    121     v2 = ( randu()*100.0 ) + 50.0;
    122     accumulator( v1, v2 );
    123 }
    124 console.log( accumulator() );
    125 ```
    126 
    127 </section>
    128 
    129 <!-- /.examples -->
    130 
    131 <section class="links">
    132 
    133 [mean-absolute-percentage-error]: https://en.wikipedia.org/wiki/Mean_absolute_percentage_error
    134 
    135 </section>
    136 
    137 <!-- /.links -->