feat: add stats/incr/incrnanmae

lib/node_modules/@stdlib/stats/incr/nanmae/README.md

@@ -0,0 +1,158 @@
+<!--
+
+@license Apache-2.0
+
+Copyright (c) 2026 The Stdlib Authors.
+
+Licensed under the Apache License, Version 2.0 (the "License");
+you may not use this file except in compliance with the License.
+You may obtain a copy of the License at
+
+   http://www.apache.org/licenses/LICENSE-2.0
+
+Unless required by applicable law or agreed to in writing, software
+distributed under the License is distributed on an "AS IS" BASIS,
+WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+See the License for the specific language governing permissions and
+limitations under the License.
+
+-->
+
+# incrnanmae
+
+> Compute the [mean absolute error][mean-absolute-error] (MAE) incrementally, ignoring `NaN` values.
+
+<section class="intro">
+
+The [mean absolute error][mean-absolute-error] is defined as
+
+<!-- <equation class="equation" label="eq:mean_absolute_error" align="center" raw="\operatorname{MAE}  = \frac{\displaystyle\sum_{i=0}^{n-1} |y_i - x_i|}{n}" alt="Equation for the mean absolute error."> -->
+
+```math
+\mathop{\mathrm{MAE}}  = \frac{\displaystyle\sum_{i=0}^{n-1} |y_i - x_i|}{n}
+```
+
+<!-- <div class="equation" align="center" data-raw-text="\operatorname{MAE}  = \frac{\displaystyle\sum_{i=0}^{n-1} |y_i - x_i|}{n}" data-equation="eq:mean_absolute_error">
+    <img src="https://cdn.jsdelivr.net/gh/stdlib-js/stdlib@49d8cabda84033d55d7b8069f19ee3dd8b8d1496/lib/node_modules/@stdlib/stats/incr/mae/docs/img/equation_mean_absolute_error.svg" alt="Equation for the mean absolute error.">
+    <br>
+</div> -->
+
+<!-- </equation> -->
+
+</section>
+
+<!-- /.intro -->
+
+<section class="usage">
+
+## Usage
+
+```javascript
+var incrnanmae = require( '@stdlib/stats/incr/nanmae' );
+```
+
+#### incrnanmae()
+
+Returns an accumulator `function` which incrementally computes the [mean absolute error][mean-absolute-error] ignoring NaN values.
+
+```javascript
+var accumulator = incrnanmae();
+```
+
+#### accumulator( \[x, y] )
+
+If provided input values `x` and `y`, the accumulator function returns an updated [mean absolute error][mean-absolute-error]. If not provided input values `x` and `y`, the accumulator function returns the current [mean absolute error][mean-absolute-error].
+
+```javascript
+var accumulator = incrnanmae();
+
+var m = accumulator();
+// returns null
+
+m = accumulator( 2.0, 3.0 );
+// returns 1.0
+
+m = accumulator( NaN, 2.0 );
+// returns 1.0
+
+m = accumulator( -5.0, NaN );
+// returns 1.0
+
+m = accumulator( NaN, NaN );
+// returns 1.0
+
+m = accumulator( -5.0, 2.0 );
+// returns 4.0
+
+m = accumulator();
+// returns 4.0
+```
+
+</section>
+
+<!-- /.usage -->
+
+<section class="notes">
+
+## Notes
+
+-   Input values are not type checked beyond checking for NaN. If provided values which are not numbers or NaN (e.g., strings, booleans), the behavior of the accumulator is undefined. If non-numeric inputs are possible, you are advised to type check and handle accordingly before passing the value to the accumulator function.
+-   If both provided values are NaN, the accumulated value is unchanged. The accumulator only updates when both inputs are valid numbers.
+-   **Warning**: the [mean absolute error][mean-absolute-error] is scale-dependent and, thus, the measure should **not** be used to make comparisons between datasets having different scales.
+
+</section>
+
+<!-- /.notes -->
+
+<section class="examples">
+
+## Examples
+
+<!-- eslint no-undef: "error" -->
+
+```javascript
+var randu = require( '@stdlib/random/base/randu' );
+var incrnanmae = require( '@stdlib/stats/incr/nanmae' );
+
+var accumulator;
+var v1;
+var v2;
+var i;
+
+// Initialize an accumulator:
+accumulator = incrnanmae();
+
+// For each simulated datum, update the mean absolute error...
+for ( i = 0; i < 100; i++ ) {
+    v1 = ( randu() < 0.2 ) ? NaN : ( randu()*100.0 ) - 50.0;
+    v2 = ( randu() < 0.2 ) ? NaN : ( randu()*100.0 ) - 50.0;
+    accumulator( v1, v2 );
+}
+console.log( accumulator() );
+```
+
+</section>
+
+<!-- /.examples -->
+
+<!-- Section for related `stdlib` packages. Do not manually edit this section, as it is automatically populated. -->
+
+<section class="related">
+
+</section>
+
+<!-- /.related -->
+
+<!-- Section for all links. Make sure to keep an empty line after the `section` element and another before the `/section` close. -->
+
+<section class="links">
+
+[mean-absolute-error]: https://en.wikipedia.org/wiki/Mean_absolute_error
+
+<!-- <related-links> -->
+
+<!-- </related-links> -->
+
+</section>
+
+<!-- /.links -->

lib/node_modules/@stdlib/stats/incr/nanmae/benchmark/benchmark.js

@@ -0,0 +1,106 @@
+/**
+* @license Apache-2.0
+*
+* Copyright (c) 2026 The Stdlib Authors.
+*
+* Licensed under the Apache License, Version 2.0 (the "License");
+* you may not use this file except in compliance with the License.
+* You may obtain a copy of the License at
+*
+*    http://www.apache.org/licenses/LICENSE-2.0
+*
+* Unless required by applicable law or agreed to in writing, software
+* distributed under the License is distributed on an "AS IS" BASIS,
+* WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+* See the License for the specific language governing permissions and
+* limitations under the License.
+*/
+
+'use strict';
+
+// MODULES //
+
+var bench = require( '@stdlib/bench' );
+var randu = require( '@stdlib/random/base/randu' );
+var isnan = require( '@stdlib/math/base/assert/is-nan' );
+var format = require( '@stdlib/string/format' );
+var pkg = require( './../package.json' ).name;
+var incrnanmae = require( './../lib' );
+
+
+// MAIN //
+
+bench( pkg, function benchmark( b ) {
+	var f;
+	var i;
+	b.tic();
+	for ( i = 0; i < b.iterations; i++ ) {
+		f = incrnanmae();
+		if ( typeof f !== 'function' ) {
+			b.fail( 'should return a function' );
+		}
+	}
+	b.toc();
+	if ( typeof f !== 'function' ) {
+		b.fail( 'should return a function' );
+	}
+	b.pass( 'benchmark finished' );
+	b.end();
+});
+
+bench( format( '%s::accumulator', pkg ), function benchmark( b ) {
+	var acc;
+	var v;
+	var i;
+	var x;
+	var y;
+
+	acc = incrnanmae();
+
+	b.tic();
+	for ( i = 0; i < b.iterations; i++ ) {
+		x = ( randu() < 0.2 ) ? NaN : randu()-0.5;
+		y = ( randu() < 0.2 ) ? NaN : randu()-0.5;
+		v = acc( x, y );
+		if ( v !== v && !isnan( x ) && !isnan( y ) ) {
+			b.fail( 'should not return NaN when inputs are not NaN' );
+		}
+	}
+	b.toc();
+	if ( v !== v && !isnan( x ) && !isnan( y ) ) {
+		b.fail( 'should not return NaN when inputs are not NaN' );
+	}
+	b.pass( 'benchmark finished' );
+	b.end();
+});
+
+bench( format( '%s::accumulator,nan_handling', pkg ), function benchmark( b ) {
+	var acc;
+	var v;
+	var i;
+
+	b.tic();
+	for ( i = 0; i < b.iterations; i++ ) {
+		acc = incrnanmae(); // Reset accumulator for each iteration
+		v = acc(); // Initial state
+		if ( v !== null ) {
+			b.fail( 'initial value should be null' );
+		}
+		acc( 2.0, 3.0 ); // Set a baseline MAE
+		v = acc( NaN, randu()-0.5 );
+		if ( typeof v !== 'number' || isnan( v ) ) {
+			b.fail( 'should return a number when one input is NaN' );
+		}
+		v = acc( randu()-0.5, NaN );
+		if ( typeof v !== 'number' || isnan( v ) ) {
+			b.fail( 'should return a number when one input is NaN' );
+		}
+		v = acc( NaN, NaN );
+		if ( typeof v !== 'number' || isnan( v ) ) {
+			b.fail( 'should return a number when both inputs are NaN' );
+		}
+	}
+	b.toc();
+	b.pass( 'benchmark finished' );
+	b.end();
+});

lib/node_modules/@stdlib/stats/incr/nanmae/docs/repl.txt

@@ -0,0 +1,36 @@
+
+{{alias}}()
+    Returns an accumulator function which incrementally computes the mean
+    absolute error, ignoring `NaN` values.
+
+    If provided two values, the accumulator function returns an updated mean
+    absolute error. If not provided values, the accumulator function returns
+    the current mean absolute error.
+
+    Returns
+    -------
+    acc: Function
+        Accumulator function.
+
+    Examples
+    --------
+    > var incrnanmae = require( '@stdlib/stats/incr/nanmae' );
+    > var accumulator = incrnanmae();
+    > var m = accumulator()
+    null
+    > m = accumulator( 2.0, 3.0 )
+    1.0
+    > m = accumulator( NaN, 2.0 )
+    1.0
+    > m = accumulator( -5.0, NaN )
+    1.0
+    > m = accumulator( NaN, NaN )
+    1.0
+    > m = accumulator( -5.0, 2.0 )
+    4.0
+    > m = accumulator()
+    4.0
+
+    See Also
+    --------
+