import clone from 'just-clone';
-// import { Constants } from './internal';
import { Constants } from './Constants';
import { WebSocketCloseEventStatusString } from '../types';
return value == null;
}
- public static isEmptyArray(object: unknown | unknown[]): boolean {
+ public static isEmptyArray(object: unknown): boolean {
return Array.isArray(object) && object.length === 0;
}
- public static isNotEmptyArray(object: unknown | unknown[]): boolean {
+ public static isNotEmptyArray(object: unknown): boolean {
return Array.isArray(object) && object.length > 0;
}
}
return '(Unknown)';
}
+
+ public static median(dataSet: number[]): number {
+ if (Utils.isEmptyArray(dataSet)) {
+ return 0;
+ }
+ if (Array.isArray(dataSet) === true && dataSet.length === 1) {
+ return dataSet[0];
+ }
+ const sortedDataSet = dataSet.slice().sort((a, b) => a - b);
+ return (
+ (sortedDataSet[(sortedDataSet.length - 1) >> 1] + sortedDataSet[sortedDataSet.length >> 1]) /
+ 2
+ );
+ }
+
+ // TODO: use order statistics tree https://en.wikipedia.org/wiki/Order_statistic_tree
+ public static percentile(dataSet: number[], percentile: number): number {
+ if (percentile < 0 && percentile > 100) {
+ throw new RangeError('Percentile is not between 0 and 100');
+ }
+ if (Utils.isEmptyArray(dataSet)) {
+ return 0;
+ }
+ const sortedDataSet = dataSet.slice().sort((a, b) => a - b);
+ if (percentile === 0 || sortedDataSet.length === 1) {
+ return sortedDataSet[0];
+ }
+ if (percentile === 100) {
+ return sortedDataSet[sortedDataSet.length - 1];
+ }
+ const percentileIndexBase = (percentile / 100) * (sortedDataSet.length - 1);
+ const percentileIndexInteger = Math.floor(percentileIndexBase);
+ if (!Utils.isNullOrUndefined(sortedDataSet[percentileIndexInteger + 1])) {
+ return (
+ sortedDataSet[percentileIndexInteger] +
+ (percentileIndexBase - percentileIndexInteger) *
+ (sortedDataSet[percentileIndexInteger + 1] - sortedDataSet[percentileIndexInteger])
+ );
+ }
+ return sortedDataSet[percentileIndexInteger];
+ }
+
+ public static stdDeviation(dataSet: number[]): number {
+ let totalDataSet = 0;
+ for (const data of dataSet) {
+ totalDataSet += data;
+ }
+ const dataSetMean = totalDataSet / dataSet.length;
+ let totalGeometricDeviation = 0;
+ for (const data of dataSet) {
+ const deviation = data - dataSetMean;
+ totalGeometricDeviation += deviation * deviation;
+ }
+ return Math.sqrt(totalGeometricDeviation / dataSet.length);
+ }
}