@visactor/vmind/cjs/atom/dataInsight/algorithms/statistics.js

<div align="center"> <a href="https://github.com/VisActor#gh-light-mode-only" target="_blank"> <img alt="VisActor Logo" width="200" src="https://github.com/VisActor/.github/blob/main/profile/logo_500_200_light.svg"/> </a> <a href="https://githu

"use strict";

var __importDefault = this && this.__importDefault || function(mod) {
    return mod && mod.__esModule ? mod : {
        default: mod
    };
};

Object.defineProperty(exports, "__esModule", {
    value: !0
}), exports.getMeanAndstdDev = exports.longestTrendInterval = exports.coefficientVariation = exports.studentTQuantile = exports.pearsonCorrelationCoeff = exports.spearmanCoefficient = exports.originalMKTest = exports.TrendType = void 0;

const jstat_1 = __importDefault(require("jstat")), stats_base_dists_t_quantile_1 = __importDefault(require("@stdlib/stats-base-dists-t-quantile"));

var TrendType;

function mkScore(x, n) {
    let s = 0;
    for (let k = 0; k < n - 1; k++) for (let j = k + 1; j < n; j++) x[j] > x[k] ? s += 1 : x[j] < x[k] && (s -= 1);
    return s;
}

function varianceS(x, n) {
    const uniqueX = Array.from(new Set(x)), g = uniqueX.length;
    let varS;
    if (n === g) varS = n * (n - 1) * (2 * n + 5) / 18; else {
        const tp = new Array(g).fill(0);
        for (let i = 0; i < g; i++) tp[i] = x.filter((item => item === uniqueX[i])).length;
        varS = (n * (n - 1) * (2 * n + 5) - tp.reduce(((acc, curr) => acc + curr * (curr - 1) * (2 * curr + 5)), 0)) / 18;
    }
    return varS;
}

function zScore(s, varS) {
    let z;
    return s > 0 ? z = (s - 1) / Math.sqrt(varS) : 0 === s ? z = 0 : s < 0 && (z = (s + 1) / Math.sqrt(varS)), 
    z;
}

function sensEstimator(x) {
    let idx = 0;
    const n = x.length, d = new Array(n * (n - 1) / 2).fill(1);
    for (let i = 0; i < n - 1; i++) for (let j = i + 1; j < n; j++) d[idx] = (x[j] - x[i]) / (j - i), 
    idx++;
    return d;
}

function median(array) {
    const sortedArray = array.filter((item => !isNaN(item))).sort(((a, b) => a - b)), midIndex = Math.floor(sortedArray.length / 2);
    return sortedArray.length % 2 == 0 ? (sortedArray[midIndex - 1] + sortedArray[midIndex]) / 2 : sortedArray[midIndex];
}

function sensSlope(x) {
    const n = x.length, slope = median(sensEstimator(x));
    return {
        slope: slope,
        intercept: median(x) - median(Array.from(Array(n).keys())) * slope
    };
}

function pValue(z, alpha) {
    const p = 2 * (1 - jstat_1.default.normal.cdf(Math.abs(z), 0, 1)), h = Math.abs(z) > jstat_1.default.normal.inv(1 - alpha / 2, 0, 1);
    let trend;
    return trend = z < 0 && h ? TrendType.DECREASING : z > 0 && h ? TrendType.INCREASING : TrendType.NO_TREND, 
    {
        p: p,
        h: h,
        trend: trend
    };
}

function originalMKTest(xOld, alpha = .05, calcScope = !1) {
    const x = xOld, n = x.length, s = mkScore(x, n), z = zScore(s, varianceS(x, n)), {p: p, h: h, trend: trend} = pValue(z, alpha);
    let finalSlope, finalIntercept;
    if (h && calcScope) {
        const {slope: slope, intercept: intercept} = sensSlope(xOld);
        finalSlope = slope, finalIntercept = intercept;
    }
    return {
        trend: trend,
        pValue: p,
        zScore: z,
        slope: finalSlope,
        intercept: finalIntercept
    };
}

function spearmanCoefficient(array1, array2) {
    return jstat_1.default.spearmancoeff(array1, array2);
}

function pearsonCorrelationCoeff(array1, array2) {
    return jstat_1.default.corrcoeff(array1, array2);
}

function studentTQuantile(x, degree) {
    return (0, stats_base_dists_t_quantile_1.default)(x, degree);
}

function coefficientVariation(data) {
    return jstat_1.default.coeffvar(data);
}

function longestTrendInterval(data, trendType) {
    if (0 === data.length) return {
        length: 0,
        start: -1,
        end: -1
    };
    let maxLength = 1, currentLength = 1, start = 0, end = 0, maxStart = 0, maxEnd = 0, maxTrend = 0, trend = 0;
    for (let i = 1; i <= data.length; i++) {
        const currentTrend = i === data.length ? null : Math.sign(data[i] - data[i - 1]);
        currentTrend === trend ? (currentLength++, end = i) : (currentLength > maxLength && (!trendType || (0 === (trendValue = trend) ? TrendType.NO_TREND : 1 === trendValue ? TrendType.INCREASING : TrendType.DECREASING) === trendType) && (maxLength = currentLength, 
        maxStart = start, maxEnd = end, maxTrend = trend), trend = currentTrend, currentLength = 0 !== currentTrend ? 2 : 1, 
        start = i - 1, end = i);
    }
    var trendValue;
    return {
        length: maxLength,
        start: maxStart,
        end: maxEnd,
        maxTrend: maxTrend
    };
}

!function(TrendType) {
    TrendType.NO_TREND = "no trend", TrendType.INCREASING = "increasing", TrendType.DECREASING = "decreasing";
}(TrendType = exports.TrendType || (exports.TrendType = {})), exports.originalMKTest = originalMKTest, 
exports.spearmanCoefficient = spearmanCoefficient, exports.pearsonCorrelationCoeff = pearsonCorrelationCoeff, 
exports.studentTQuantile = studentTQuantile, exports.coefficientVariation = coefficientVariation, 
exports.longestTrendInterval = longestTrendInterval;

const getMeanAndstdDev = data => {
    const validData = data.filter((v => !isNaN(v))), mean = validData.reduce(((sum, v) => sum + v), 0) / validData.length, stdDev = Math.sqrt(validData.reduce(((sum, v) => sum + Math.pow(v - mean, 2)), 0) / validData.length);
    return {
        mean: mean,
        stdDev: stdDev,
        max: Math.max(...validData),
        min: Math.min(...validData)
    };
};

exports.getMeanAndstdDev = getMeanAndstdDev;
//# sourceMappingURL=statistics.js.map