@visactor/vutils 1.0.24 → 1.0.26

package/es/common/regression-linear.d.ts

@@ -1,3 +1,7 @@
+/**
+ * Linear regression utilities (single clean implementation).
+ * Exports: ordinaryLeastSquares, visitPoints, rSquared, regressionLinear
+ */
 export declare function ordinaryLeastSquares(uX: number, uY: number, uXY: number, uX2: number): {
     a: number;
     b: number;
@@ -11,6 +15,7 @@ export declare function regressionLinear(data: any[], x?: (d: any) => number, y?
         a: number;
         b: number;
     };
+    rank: number;
     predict: (xx: number) => number;
     rSquared: number;
     evaluateGrid: (N: number) => {

package/es/common/regression-linear.js

@@ -1,8 +1,13 @@
 import isNil from "./isNil";
 
-import { computeLinearCIComponents, invNorm, stdErrorsAt } from "./regression-utils";
+import { regressionPoints, fitRegression, regressionPrediction, regressionCoefficients, regressionGrid } from "./regression-solver";
 
-export function ordinaryLeastSquares(uX, uY, uXY, uX2) {
+import { computeLinearCIComponents, confidenceCriticalValue, confidenceHalfWidth, stdErrorsAt } from "./regression-utils";
+
+/**
+ * Linear regression utilities (single clean implementation).
+ * Exports: ordinaryLeastSquares, visitPoints, rSquared, regressionLinear
+ */ export function ordinaryLeastSquares(uX, uY, uXY, uX2) {
     const denom = uX2 - uX * uX;
     if (Math.abs(denom) < Number.EPSILON) return {
         a: uY,
@@ -40,66 +45,47 @@ export function rSquared(data, x, y, uY, predict) {
 
 export function regressionLinear(data, x = (d => d.x), y = (d => d.y), options) {
     var _a;
-    const alpha = null !== (_a = null == options ? void 0 : options.alpha) && void 0 !== _a ? _a : .05;
-    let n = 0, meanX = 0, meanY = 0, meanXY = 0, meanX2 = 0;
-    visitPoints(data, x, y, ((xi, yi) => {
-        n++, meanX += (xi - meanX) / n, meanY += (yi - meanY) / n, meanXY += (xi * yi - meanXY) / n, 
-        meanX2 += (xi * xi - meanX2) / n;
-    }));
-    const {a: a, b: b} = ordinaryLeastSquares(meanX, meanY, meanXY, meanX2), predict = xx => a + b * xx, comps = computeLinearCIComponents(data, x, y, predict);
-    return {
+    const alpha = null !== (_a = null == options ? void 0 : options.alpha) && void 0 !== _a ? _a : .05, points = regressionPoints(data, x, y), model = fitRegression(points, 1), coef = regressionCoefficients(model), a = coef[0], b = coef[1], predict = xx => regressionPrediction(model, xx), comps = computeLinearCIComponents(data, x, y, predict);
+    let meanY = 0;
+    return points.forEach(((p, i) => {
+        meanY += (p.y - meanY) / (i + 1);
+    })), {
         coef: {
             a: a,
             b: b
         },
+        rank: model.rank,
         predict: predict,
         rSquared: rSquared(data, x, y, meanY, predict),
         evaluateGrid: function(N) {
-            const out = [];
-            if (0 === comps.n || N <= 0) return out;
-            if (comps.min === comps.max) {
-                for (let i = 0; i < N; i++) out.push({
-                    x: comps.min,
-                    y: predict(comps.min)
-                });
-                return out;
-            }
-            const step = (comps.max - comps.min) / (N - 1);
-            for (let i = 0; i < N; i++) {
-                const px = i === N - 1 ? comps.max : comps.min + step * i;
-                out.push({
-                    x: px,
-                    y: predict(px)
-                });
-            }
-            return out;
+            return regressionGrid(comps.min, comps.max, N, predict);
         },
         confidenceInterval: function(N = 50) {
             const out = [];
             if (0 === comps.n || N <= 0) return out;
-            const z = invNorm(1 - alpha / 2);
+            const z = confidenceCriticalValue(alpha);
             if (comps.min === comps.max) {
-                const m = predict(comps.min), errs = stdErrorsAt(comps.min, comps);
+                const m = predict(comps.min), errs = stdErrorsAt(comps.min, comps), meanHalfWidth = confidenceHalfWidth(z, errs.seMean), predHalfWidth = confidenceHalfWidth(z, errs.sePred);
                 for (let i = 0; i < N; i++) out.push({
                     x: comps.min,
                     mean: m,
-                    lower: m - z * errs.seMean,
-                    upper: m + z * errs.seMean,
-                    predLower: m - z * errs.sePred,
-                    predUpper: m + z * errs.sePred
+                    lower: m - meanHalfWidth,
+                    upper: m + meanHalfWidth,
+                    predLower: m - predHalfWidth,
+                    predUpper: m + predHalfWidth
                 });
                 return out;
             }
             const step = (comps.max - comps.min) / (N - 1);
             for (let i = 0; i < N; i++) {
-                const px = i === N - 1 ? comps.max : comps.min + step * i, m = predict(px), errs = stdErrorsAt(px, comps);
+                const px = i === N - 1 ? comps.max : comps.min + step * i, m = predict(px), errs = stdErrorsAt(px, comps), meanHalfWidth = confidenceHalfWidth(z, errs.seMean), predHalfWidth = confidenceHalfWidth(z, errs.sePred);
                 out.push({
                     x: px,
                     mean: m,
-                    lower: m - z * errs.seMean,
-                    upper: m + z * errs.seMean,
-                    predLower: m - z * errs.sePred,
-                    predUpper: m + z * errs.sePred
+                    lower: m - meanHalfWidth,
+                    upper: m + meanHalfWidth,
+                    predLower: m - predHalfWidth,
+                    predUpper: m + predHalfWidth
                 });
             }
             return out;

package/es/common/regression-linear.js.map

@@ -1 +1 @@
-{"version":3,"sources":["../src/common/regression-linear.ts"],"names":[],"mappings":"AAAA,OAAO,KAAK,MAAM,SAAS,CAAC;AAC5B,OAAO,EAAE,yBAAyB,EAAE,OAAO,EAAE,WAAW,EAAE,MAAM,oBAAoB,CAAC;AAOrF,MAAM,UAAU,oBAAoB,CAAC,EAAU,EAAE,EAAU,EAAE,GAAW,EAAE,GAAW;IACnF,MAAM,KAAK,GAAG,GAAG,GAAG,EAAE,GAAG,EAAE,CAAC;IAC5B,IAAI,IAAI,CAAC,GAAG,CAAC,KAAK,CAAC,GAAG,MAAM,CAAC,OAAO,EAAE;QACpC,OAAO,EAAE,CAAC,EAAE,EAAE,EAAE,CAAC,EAAE,CAAC,EAAE,CAAC;KACxB;IACD,MAAM,CAAC,GAAG,CAAC,GAAG,GAAG,EAAE,GAAG,EAAE,CAAC,GAAG,KAAK,CAAC;IAClC,MAAM,CAAC,GAAG,EAAE,GAAG,CAAC,GAAG,EAAE,CAAC;IACtB,OAAO,EAAE,CAAC,EAAE,CAAC,EAAE,CAAC;AAClB,CAAC;AAED,MAAM,UAAU,WAAW,CACzB,IAAW,EACX,CAAqB,EACrB,CAAqB,EACrB,QAAuD;IAEvD,KAAK,IAAI,CAAC,GAAG,CAAC,EAAE,CAAC,GAAG,IAAI,CAAC,MAAM,EAAE,CAAC,EAAE,EAAE;QACpC,MAAM,CAAC,GAAG,IAAI,CAAC,CAAC,CAAC,CAAC;QAClB,IAAI,EAAE,GAAG,CAAC,CAAC,CAAC,CAAC,CAAC;QACd,IAAI,EAAE,GAAG,CAAC,CAAC,CAAC,CAAC,CAAC;QACd,IAAI,CAAC,KAAK,CAAC,EAAE,CAAC,IAAI,CAAC,EAAE,GAAG,CAAC,EAAE,CAAC,IAAI,EAAE,IAAI,CAAC,KAAK,CAAC,EAAE,CAAC,IAAI,CAAC,EAAE,GAAG,CAAC,EAAE,CAAC,IAAI,EAAE,EAAE;YACpE,QAAQ,CAAC,EAAE,EAAE,EAAE,EAAE,CAAC,CAAC,CAAC;SACrB;KACF;AACH,CAAC;AAED,MAAM,UAAU,QAAQ,CACtB,IAAW,EACX,CAAqB,EACrB,CAAqB,EACrB,EAAU,EACV,OAA8B;IAE9B,IAAI,GAAG,GAAG,CAAC,CAAC;IACZ,IAAI,GAAG,GAAG,CAAC,CAAC;IACZ,KAAK,IAAI,CAAC,GAAG,CAAC,EAAE,CAAC,GAAG,IAAI,CAAC,MAAM,EAAE,CAAC,EAAE,EAAE;QACpC,MAAM,CAAC,GAAG,IAAI,CAAC,CAAC,CAAC,CAAC;QAClB,IAAI,EAAE,GAAG,CAAC,CAAC,CAAC,CAAC,CAAC;QACd,IAAI,CAAC,KAAK,CAAC,EAAE,CAAC,IAAI,CAAC,EAAE,GAAG,CAAC,EAAE,CAAC,IAAI,EAAE,EAAE;YAClC,MAAM,CAAC,GAAG,OAAO,CAAC,CAAC,CAAC,CAAC,CAAC,CAAC,CAAC;YACxB,MAAM,CAAC,GAAG,EAAE,GAAG,CAAC,CAAC;YACjB,GAAG,IAAI,CAAC,GAAG,CAAC,CAAC;YACb,MAAM,CAAC,GAAG,EAAE,GAAG,EAAE,CAAC;YAClB,GAAG,IAAI,CAAC,GAAG,CAAC,CAAC;SACd;KACF;IACD,OAAO,GAAG,KAAK,CAAC,CAAC,CAAC,CAAC,CAAC,CAAC,CAAC,CAAC,CAAC,GAAG,GAAG,GAAG,GAAG,CAAC;AACvC,CAAC;AAED,MAAM,UAAU,gBAAgB,CAC9B,IAAW,EACX,IAAwB,CAAC,CAAC,EAAE,CAAC,CAAC,CAAC,CAAC,EAChC,IAAwB,CAAC,CAAC,EAAE,CAAC,CAAC,CAAC,CAAC,EAChC,OAEC;;IAED,MAAM,KAAK,GAAG,MAAA,OAAO,aAAP,OAAO,uBAAP,OAAO,CAAE,KAAK,mCAAI,IAAI,CAAC;IAErC,IAAI,CAAC,GAAG,CAAC,CAAC;IACV,IAAI,KAAK,GAAG,CAAC,CAAC;IACd,IAAI,KAAK,GAAG,CAAC,CAAC;IACd,IAAI,MAAM,GAAG,CAAC,CAAC;IACf,IAAI,MAAM,GAAG,CAAC,CAAC;IAEf,WAAW,CAAC,IAAI,EAAE,CAAC,EAAE,CAAC,EAAE,CAAC,EAAE,EAAE,EAAE,EAAE,EAAE;QACjC,CAAC,EAAE,CAAC;QACJ,KAAK,IAAI,CAAC,EAAE,GAAG,KAAK,CAAC,GAAG,CAAC,CAAC;QAC1B,KAAK,IAAI,CAAC,EAAE,GAAG,KAAK,CAAC,GAAG,CAAC,CAAC;QAC1B,MAAM,IAAI,CAAC,EAAE,GAAG,EAAE,GAAG,MAAM,CAAC,GAAG,CAAC,CAAC;QACjC,MAAM,IAAI,CAAC,EAAE,GAAG,EAAE,GAAG,MAAM,CAAC,GAAG,CAAC,CAAC;IACnC,CAAC,CAAC,CAAC;IAEH,MAAM,EAAE,CAAC,EAAE,CAAC,EAAE,GAAG,oBAAoB,CAAC,KAAK,EAAE,KAAK,EAAE,MAAM,EAAE,MAAM,CAAC,CAAC;IACpE,MAAM,OAAO,GAAG,CAAC,EAAU,EAAE,EAAE,CAAC,CAAC,GAAG,CAAC,GAAG,EAAE,CAAC;IAE3C,MAAM,KAAK,GAAG,yBAAyB,CAAC,IAAI,EAAE,CAAC,EAAE,CAAC,EAAE,OAAO,CAAC,CAAC;IAE7D,SAAS,YAAY,CAAC,CAAS;QAC7B,MAAM,GAAG,GAA+B,EAAE,CAAC;QAC3C,IAAI,KAAK,CAAC,CAAC,KAAK,CAAC,IAAI,CAAC,IAAI,CAAC,EAAE;YAC3B,OAAO,GAAG,CAAC;SACZ;QACD,IAAI,KAAK,CAAC,GAAG,KAAK,KAAK,CAAC,GAAG,EAAE;YAC3B,KAAK,IAAI,CAAC,GAAG,CAAC,EAAE,CAAC,GAAG,CAAC,EAAE,CAAC,EAAE,EAAE;gBAC1B,GAAG,CAAC,IAAI,CAAC,EAAE,CAAC,EAAE,KAAK,CAAC,GAAG,EAAE,CAAC,EAAE,OAAO,CAAC,KAAK,CAAC,GAAG,CAAC,EAAE,CAAC,CAAC;aACnD;YACD,OAAO,GAAG,CAAC;SACZ;QACD,MAAM,IAAI,GAAG,CAAC,KAAK,CAAC,GAAG,GAAG,KAAK,CAAC,GAAG,CAAC,GAAG,CAAC,CAAC,GAAG,CAAC,CAAC,CAAC;QAC/C,KAAK,IAAI,CAAC,GAAG,CAAC,EAAE,CAAC,GAAG,CAAC,EAAE,CAAC,EAAE,EAAE;YAC1B,MAAM,EAAE,GAAG,CAAC,KAAK,CAAC,GAAG,CAAC,CAAC,CAAC,CAAC,KAAK,CAAC,GAAG,CAAC,CAAC,CAAC,KAAK,CAAC,GAAG,GAAG,IAAI,GAAG,CAAC,CAAC;YAC1D,GAAG,CAAC,IAAI,CAAC,EAAE,CAAC,EAAE,EAAE,EAAE,CAAC,EAAE,OAAO,CAAC,EAAE,CAAC,EAAE,CAAC,CAAC;SACrC;QACD,OAAO,GAAG,CAAC;IACb,CAAC;IAED,SAAS,kBAAkB,CAAC,IAAY,EAAE;QACxC,MAAM,GAAG,GAAsG,EAAE,CAAC;QAClH,IAAI,KAAK,CAAC,CAAC,KAAK,CAAC,IAAI,CAAC,IAAI,CAAC,EAAE;YAC3B,OAAO,GAAG,CAAC;SACZ;QACD,MAAM,CAAC,GAAG,OAAO,CAAC,CAAC,GAAG,KAAK,GAAG,CAAC,CAAC,CAAC;QACjC,IAAI,KAAK,CAAC,GAAG,KAAK,KAAK,CAAC,GAAG,EAAE;YAC3B,MAAM,CAAC,GAAG,OAAO,CAAC,KAAK,CAAC,GAAG,CAAC,CAAC;YAC7B,MAAM,IAAI,GAAG,WAAW,CAAC,KAAK,CAAC,GAAG,EAAE,KAAK,CAAC,CAAC;YAC3C,KAAK,IAAI,CAAC,GAAG,CAAC,EAAE,CAAC,GAAG,CAAC,EAAE,CAAC,EAAE,EAAE;gBAC1B,GAAG,CAAC,IAAI,CAAC;oBACP,CAAC,EAAE,KAAK,CAAC,GAAG;oBACZ,IAAI,EAAE,CAAC;oBACP,KAAK,EAAE,CAAC,GAAG,CAAC,GAAG,IAAI,CAAC,MAAM;oBAC1B,KAAK,EAAE,CAAC,GAAG,CAAC,GAAG,IAAI,CAAC,MAAM;oBAC1B,SAAS,EAAE,CAAC,GAAG,CAAC,GAAG,IAAI,CAAC,MAAM;oBAC9B,SAAS,EAAE,CAAC,GAAG,CAAC,GAAG,IAAI,CAAC,MAAM;iBAC/B,CAAC,CAAC;aACJ;YACD,OAAO,GAAG,CAAC;SACZ;QACD,MAAM,IAAI,GAAG,CAAC,KAAK,CAAC,GAAG,GAAG,KAAK,CAAC,GAAG,CAAC,GAAG,CAAC,CAAC,GAAG,CAAC,CAAC,CAAC;QAC/C,KAAK,IAAI,CAAC,GAAG,CAAC,EAAE,CAAC,GAAG,CAAC,EAAE,CAAC,EAAE,EAAE;YAC1B,MAAM,EAAE,GAAG,CAAC,KAAK,CAAC,GAAG,CAAC,CAAC,CAAC,CAAC,KAAK,CAAC,GAAG,CAAC,CAAC,CAAC,KAAK,CAAC,GAAG,GAAG,IAAI,GAAG,CAAC,CAAC;YAC1D,MAAM,CAAC,GAAG,OAAO,CAAC,EAAE,CAAC,CAAC;YACtB,MAAM,IAAI,GAAG,WAAW,CAAC,EAAE,EAAE,KAAK,CAAC,CAAC;YACpC,GAAG,CAAC,IAAI,CAAC;gBACP,CAAC,EAAE,EAAE;gBACL,IAAI,EAAE,CAAC;gBACP,KAAK,EAAE,CAAC,GAAG,CAAC,GAAG,IAAI,CAAC,MAAM;gBAC1B,KAAK,EAAE,CAAC,GAAG,CAAC,GAAG,IAAI,CAAC,MAAM;gBAC1B,SAAS,EAAE,CAAC,GAAG,CAAC,GAAG,IAAI,CAAC,MAAM;gBAC9B,SAAS,EAAE,CAAC,GAAG,CAAC,GAAG,IAAI,CAAC,MAAM;aAC/B,CAAC,CAAC;SACJ;QACD,OAAO,GAAG,CAAC;IACb,CAAC;IAED,OAAO;QACL,IAAI,EAAE,EAAE,CAAC,EAAE,CAAC,EAAE;QACd,OAAO;QACP,QAAQ,EAAE,QAAQ,CAAC,IAAI,EAAE,CAAC,EAAE,CAAC,EAAE,KAAK,EAAE,OAAO,CAAC;QAC9C,YAAY;QACZ,kBAAkB;KACnB,CAAC;AACJ,CAAC;AAED,eAAe;IACb,oBAAoB;IACpB,WAAW;IACX,QAAQ;IACR,gBAAgB;CACjB,CAAC","file":"regression-linear.js","sourcesContent":["import isNil from './isNil';\nimport { computeLinearCIComponents, invNorm, stdErrorsAt } from './regression-utils';\n\n/**\n * Linear regression utilities (single clean implementation).\n * Exports: ordinaryLeastSquares, visitPoints, rSquared, regressionLinear\n */\n\nexport function ordinaryLeastSquares(uX: number, uY: number, uXY: number, uX2: number) {\n  const denom = uX2 - uX * uX;\n  if (Math.abs(denom) < Number.EPSILON) {\n    return { a: uY, b: 0 };\n  }\n  const b = (uXY - uX * uY) / denom;\n  const a = uY - b * uX;\n  return { a, b };\n}\n\nexport function visitPoints(\n  data: any[],\n  x: (d: any) => number,\n  y: (d: any) => number,\n  callback: (x: number, y: number, index: number) => void\n) {\n  for (let i = 0; i < data.length; i++) {\n    const d = data[i];\n    let xi = x(d);\n    let yi = y(d);\n    if (!isNil(xi) && (xi = +xi) >= xi && !isNil(yi) && (yi = +yi) >= yi) {\n      callback(xi, yi, i);\n    }\n  }\n}\n\nexport function rSquared(\n  data: any[],\n  x: (d: any) => number,\n  y: (d: any) => number,\n  uY: number,\n  predict: (x: number) => number\n) {\n  let ssr = 0;\n  let sst = 0;\n  for (let i = 0; i < data.length; i++) {\n    const d = data[i];\n    let yi = y(d);\n    if (!isNil(yi) && (yi = +yi) >= yi) {\n      const p = predict(x(d));\n      const r = yi - p;\n      ssr += r * r;\n      const t = yi - uY;\n      sst += t * t;\n    }\n  }\n  return sst === 0 ? 0 : 1 - ssr / sst;\n}\n\nexport function regressionLinear(\n  data: any[],\n  x: (d: any) => number = d => d.x,\n  y: (d: any) => number = d => d.y,\n  options?: {\n    alpha?: number;\n  }\n) {\n  const alpha = options?.alpha ?? 0.05;\n  // accumulate online means (sufficient statistics)\n  let n = 0;\n  let meanX = 0;\n  let meanY = 0;\n  let meanXY = 0;\n  let meanX2 = 0;\n\n  visitPoints(data, x, y, (xi, yi) => {\n    n++;\n    meanX += (xi - meanX) / n;\n    meanY += (yi - meanY) / n;\n    meanXY += (xi * yi - meanXY) / n;\n    meanX2 += (xi * xi - meanX2) / n;\n  });\n\n  const { a, b } = ordinaryLeastSquares(meanX, meanY, meanXY, meanX2);\n  const predict = (xx: number) => a + b * xx;\n\n  const comps = computeLinearCIComponents(data, x, y, predict);\n\n  function evaluateGrid(N: number) {\n    const out: { x: number; y: number }[] = [];\n    if (comps.n === 0 || N <= 0) {\n      return out;\n    }\n    if (comps.min === comps.max) {\n      for (let i = 0; i < N; i++) {\n        out.push({ x: comps.min, y: predict(comps.min) });\n      }\n      return out;\n    }\n    const step = (comps.max - comps.min) / (N - 1);\n    for (let i = 0; i < N; i++) {\n      const px = i === N - 1 ? comps.max : comps.min + step * i;\n      out.push({ x: px, y: predict(px) });\n    }\n    return out;\n  }\n\n  function confidenceInterval(N: number = 50) {\n    const out: { x: number; mean: number; lower: number; upper: number; predLower: number; predUpper: number }[] = [];\n    if (comps.n === 0 || N <= 0) {\n      return out;\n    }\n    const z = invNorm(1 - alpha / 2);\n    if (comps.min === comps.max) {\n      const m = predict(comps.min);\n      const errs = stdErrorsAt(comps.min, comps);\n      for (let i = 0; i < N; i++) {\n        out.push({\n          x: comps.min,\n          mean: m,\n          lower: m - z * errs.seMean,\n          upper: m + z * errs.seMean,\n          predLower: m - z * errs.sePred,\n          predUpper: m + z * errs.sePred\n        });\n      }\n      return out;\n    }\n    const step = (comps.max - comps.min) / (N - 1);\n    for (let i = 0; i < N; i++) {\n      const px = i === N - 1 ? comps.max : comps.min + step * i;\n      const m = predict(px);\n      const errs = stdErrorsAt(px, comps);\n      out.push({\n        x: px,\n        mean: m,\n        lower: m - z * errs.seMean,\n        upper: m + z * errs.seMean,\n        predLower: m - z * errs.sePred,\n        predUpper: m + z * errs.sePred\n      });\n    }\n    return out;\n  }\n\n  return {\n    coef: { a, b },\n    predict,\n    rSquared: rSquared(data, x, y, meanY, predict),\n    evaluateGrid,\n    confidenceInterval\n  };\n}\n\nexport default {\n  ordinaryLeastSquares,\n  visitPoints,\n  rSquared,\n  regressionLinear\n};\n"]}
+{"version":3,"sources":["../src/common/regression-linear.ts"],"names":[],"mappings":"AAAA,OAAO,KAAK,MAAM,SAAS,CAAC;AAC5B,OAAO,EACL,gBAAgB,EAChB,aAAa,EACb,oBAAoB,EACpB,sBAAsB,EACtB,cAAc,EACf,MAAM,qBAAqB,CAAC;AAC7B,OAAO,EACL,yBAAyB,EACzB,uBAAuB,EACvB,mBAAmB,EACnB,WAAW,EACZ,MAAM,oBAAoB,CAAC;AAE5B;;;GAGG;AAEH,MAAM,UAAU,oBAAoB,CAAC,EAAU,EAAE,EAAU,EAAE,GAAW,EAAE,GAAW;IACnF,MAAM,KAAK,GAAG,GAAG,GAAG,EAAE,GAAG,EAAE,CAAC;IAC5B,IAAI,IAAI,CAAC,GAAG,CAAC,KAAK,CAAC,GAAG,MAAM,CAAC,OAAO,EAAE;QACpC,OAAO,EAAE,CAAC,EAAE,EAAE,EAAE,CAAC,EAAE,CAAC,EAAE,CAAC;KACxB;IACD,MAAM,CAAC,GAAG,CAAC,GAAG,GAAG,EAAE,GAAG,EAAE,CAAC,GAAG,KAAK,CAAC;IAClC,MAAM,CAAC,GAAG,EAAE,GAAG,CAAC,GAAG,EAAE,CAAC;IACtB,OAAO,EAAE,CAAC,EAAE,CAAC,EAAE,CAAC;AAClB,CAAC;AAED,MAAM,UAAU,WAAW,CACzB,IAAW,EACX,CAAqB,EACrB,CAAqB,EACrB,QAAuD;IAEvD,KAAK,IAAI,CAAC,GAAG,CAAC,EAAE,CAAC,GAAG,IAAI,CAAC,MAAM,EAAE,CAAC,EAAE,EAAE;QACpC,MAAM,CAAC,GAAG,IAAI,CAAC,CAAC,CAAC,CAAC;QAClB,IAAI,EAAE,GAAG,CAAC,CAAC,CAAC,CAAC,CAAC;QACd,IAAI,EAAE,GAAG,CAAC,CAAC,CAAC,CAAC,CAAC;QACd,IAAI,CAAC,KAAK,CAAC,EAAE,CAAC,IAAI,CAAC,EAAE,GAAG,CAAC,EAAE,CAAC,IAAI,EAAE,IAAI,CAAC,KAAK,CAAC,EAAE,CAAC,IAAI,CAAC,EAAE,GAAG,CAAC,EAAE,CAAC,IAAI,EAAE,EAAE;YACpE,QAAQ,CAAC,EAAE,EAAE,EAAE,EAAE,CAAC,CAAC,CAAC;SACrB;KACF;AACH,CAAC;AAED,MAAM,UAAU,QAAQ,CACtB,IAAW,EACX,CAAqB,EACrB,CAAqB,EACrB,EAAU,EACV,OAA8B;IAE9B,IAAI,GAAG,GAAG,CAAC,CAAC;IACZ,IAAI,GAAG,GAAG,CAAC,CAAC;IACZ,KAAK,IAAI,CAAC,GAAG,CAAC,EAAE,CAAC,GAAG,IAAI,CAAC,MAAM,EAAE,CAAC,EAAE,EAAE;QACpC,MAAM,CAAC,GAAG,IAAI,CAAC,CAAC,CAAC,CAAC;QAClB,IAAI,EAAE,GAAG,CAAC,CAAC,CAAC,CAAC,CAAC;QACd,IAAI,CAAC,KAAK,CAAC,EAAE,CAAC,IAAI,CAAC,EAAE,GAAG,CAAC,EAAE,CAAC,IAAI,EAAE,EAAE;YAClC,MAAM,CAAC,GAAG,OAAO,CAAC,CAAC,CAAC,CAAC,CAAC,CAAC,CAAC;YACxB,MAAM,CAAC,GAAG,EAAE,GAAG,CAAC,CAAC;YACjB,GAAG,IAAI,CAAC,GAAG,CAAC,CAAC;YACb,MAAM,CAAC,GAAG,EAAE,GAAG,EAAE,CAAC;YAClB,GAAG,IAAI,CAAC,GAAG,CAAC,CAAC;SACd;KACF;IACD,OAAO,GAAG,KAAK,CAAC,CAAC,CAAC,CAAC,CAAC,CAAC,CAAC,CAAC,CAAC,GAAG,GAAG,GAAG,GAAG,CAAC;AACvC,CAAC;AAED,MAAM,UAAU,gBAAgB,CAC9B,IAAW,EACX,IAAwB,CAAC,CAAC,EAAE,CAAC,CAAC,CAAC,CAAC,EAChC,IAAwB,CAAC,CAAC,EAAE,CAAC,CAAC,CAAC,CAAC,EAChC,OAEC;;IAED,MAAM,KAAK,GAAG,MAAA,OAAO,aAAP,OAAO,uBAAP,OAAO,CAAE,KAAK,mCAAI,IAAI,CAAC;IACrC,MAAM,MAAM,GAAG,gBAAgB,CAAC,IAAI,EAAE,CAAC,EAAE,CAAC,CAAC,CAAC;IAC5C,MAAM,KAAK,GAAG,aAAa,CAAC,MAAM,EAAE,CAAC,CAAC,CAAC;IACvC,MAAM,IAAI,GAAG,sBAAsB,CAAC,KAAK,CAAC,CAAC;IAC3C,MAAM,CAAC,GAAG,IAAI,CAAC,CAAC,CAAC,CAAC;IAClB,MAAM,CAAC,GAAG,IAAI,CAAC,CAAC,CAAC,CAAC;IAClB,MAAM,OAAO,GAAG,CAAC,EAAU,EAAE,EAAE,CAAC,oBAAoB,CAAC,KAAK,EAAE,EAAE,CAAC,CAAC;IAChE,MAAM,KAAK,GAAG,yBAAyB,CAAC,IAAI,EAAE,CAAC,EAAE,CAAC,EAAE,OAAO,CAAC,CAAC;IAC7D,IAAI,KAAK,GAAG,CAAC,CAAC;IACd,MAAM,CAAC,OAAO,CAAC,CAAC,CAAC,EAAE,CAAC,EAAE,EAAE;QACtB,KAAK,IAAI,CAAC,CAAC,CAAC,CAAC,GAAG,KAAK,CAAC,GAAG,CAAC,CAAC,GAAG,CAAC,CAAC,CAAC;IACnC,CAAC,CAAC,CAAC;IAEH,SAAS,YAAY,CAAC,CAAS;QAC7B,OAAO,cAAc,CAAC,KAAK,CAAC,GAAG,EAAE,KAAK,CAAC,GAAG,EAAE,CAAC,EAAE,OAAO,CAAC,CAAC;IAC1D,CAAC;IAED,SAAS,kBAAkB,CAAC,IAAY,EAAE;QACxC,MAAM,GAAG,GAAsG,EAAE,CAAC;QAClH,IAAI,KAAK,CAAC,CAAC,KAAK,CAAC,IAAI,CAAC,IAAI,CAAC,EAAE;YAC3B,OAAO,GAAG,CAAC;SACZ;QACD,MAAM,CAAC,GAAG,uBAAuB,CAAC,KAAK,CAAC,CAAC;QACzC,IAAI,KAAK,CAAC,GAAG,KAAK,KAAK,CAAC,GAAG,EAAE;YAC3B,MAAM,CAAC,GAAG,OAAO,CAAC,KAAK,CAAC,GAAG,CAAC,CAAC;YAC7B,MAAM,IAAI,GAAG,WAAW,CAAC,KAAK,CAAC,GAAG,EAAE,KAAK,CAAC,CAAC;YAC3C,MAAM,aAAa,GAAG,mBAAmB,CAAC,CAAC,EAAE,IAAI,CAAC,MAAM,CAAC,CAAC;YAC1D,MAAM,aAAa,GAAG,mBAAmB,CAAC,CAAC,EAAE,IAAI,CAAC,MAAM,CAAC,CAAC;YAC1D,KAAK,IAAI,CAAC,GAAG,CAAC,EAAE,CAAC,GAAG,CAAC,EAAE,CAAC,EAAE,EAAE;gBAC1B,GAAG,CAAC,IAAI,CAAC;oBACP,CAAC,EAAE,KAAK,CAAC,GAAG;oBACZ,IAAI,EAAE,CAAC;oBACP,KAAK,EAAE,CAAC,GAAG,aAAa;oBACxB,KAAK,EAAE,CAAC,GAAG,aAAa;oBACxB,SAAS,EAAE,CAAC,GAAG,aAAa;oBAC5B,SAAS,EAAE,CAAC,GAAG,aAAa;iBAC7B,CAAC,CAAC;aACJ;YACD,OAAO,GAAG,CAAC;SACZ;QACD,MAAM,IAAI,GAAG,CAAC,KAAK,CAAC,GAAG,GAAG,KAAK,CAAC,GAAG,CAAC,GAAG,CAAC,CAAC,GAAG,CAAC,CAAC,CAAC;QAC/C,KAAK,IAAI,CAAC,GAAG,CAAC,EAAE,CAAC,GAAG,CAAC,EAAE,CAAC,EAAE,EAAE;YAC1B,MAAM,EAAE,GAAG,CAAC,KAAK,CAAC,GAAG,CAAC,CAAC,CAAC,CAAC,KAAK,CAAC,GAAG,CAAC,CAAC,CAAC,KAAK,CAAC,GAAG,GAAG,IAAI,GAAG,CAAC,CAAC;YAC1D,MAAM,CAAC,GAAG,OAAO,CAAC,EAAE,CAAC,CAAC;YACtB,MAAM,IAAI,GAAG,WAAW,CAAC,EAAE,EAAE,KAAK,CAAC,CAAC;YACpC,MAAM,aAAa,GAAG,mBAAmB,CAAC,CAAC,EAAE,IAAI,CAAC,MAAM,CAAC,CAAC;YAC1D,MAAM,aAAa,GAAG,mBAAmB,CAAC,CAAC,EAAE,IAAI,CAAC,MAAM,CAAC,CAAC;YAC1D,GAAG,CAAC,IAAI,CAAC;gBACP,CAAC,EAAE,EAAE;gBACL,IAAI,EAAE,CAAC;gBACP,KAAK,EAAE,CAAC,GAAG,aAAa;gBACxB,KAAK,EAAE,CAAC,GAAG,aAAa;gBACxB,SAAS,EAAE,CAAC,GAAG,aAAa;gBAC5B,SAAS,EAAE,CAAC,GAAG,aAAa;aAC7B,CAAC,CAAC;SACJ;QACD,OAAO,GAAG,CAAC;IACb,CAAC;IAED,OAAO;QACL,IAAI,EAAE,EAAE,CAAC,EAAE,CAAC,EAAE;QACd,IAAI,EAAE,KAAK,CAAC,IAAI;QAChB,OAAO;QACP,QAAQ,EAAE,QAAQ,CAAC,IAAI,EAAE,CAAC,EAAE,CAAC,EAAE,KAAK,EAAE,OAAO,CAAC;QAC9C,YAAY;QACZ,kBAAkB;KACnB,CAAC;AACJ,CAAC;AAED,eAAe;IACb,oBAAoB;IACpB,WAAW;IACX,QAAQ;IACR,gBAAgB;CACjB,CAAC","file":"regression-linear.js","sourcesContent":["import isNil from './isNil';\nimport {\n  regressionPoints,\n  fitRegression,\n  regressionPrediction,\n  regressionCoefficients,\n  regressionGrid\n} from './regression-solver';\nimport {\n  computeLinearCIComponents,\n  confidenceCriticalValue,\n  confidenceHalfWidth,\n  stdErrorsAt\n} from './regression-utils';\n\n/**\n * Linear regression utilities (single clean implementation).\n * Exports: ordinaryLeastSquares, visitPoints, rSquared, regressionLinear\n */\n\nexport function ordinaryLeastSquares(uX: number, uY: number, uXY: number, uX2: number) {\n  const denom = uX2 - uX * uX;\n  if (Math.abs(denom) < Number.EPSILON) {\n    return { a: uY, b: 0 };\n  }\n  const b = (uXY - uX * uY) / denom;\n  const a = uY - b * uX;\n  return { a, b };\n}\n\nexport function visitPoints(\n  data: any[],\n  x: (d: any) => number,\n  y: (d: any) => number,\n  callback: (x: number, y: number, index: number) => void\n) {\n  for (let i = 0; i < data.length; i++) {\n    const d = data[i];\n    let xi = x(d);\n    let yi = y(d);\n    if (!isNil(xi) && (xi = +xi) >= xi && !isNil(yi) && (yi = +yi) >= yi) {\n      callback(xi, yi, i);\n    }\n  }\n}\n\nexport function rSquared(\n  data: any[],\n  x: (d: any) => number,\n  y: (d: any) => number,\n  uY: number,\n  predict: (x: number) => number\n) {\n  let ssr = 0;\n  let sst = 0;\n  for (let i = 0; i < data.length; i++) {\n    const d = data[i];\n    let yi = y(d);\n    if (!isNil(yi) && (yi = +yi) >= yi) {\n      const p = predict(x(d));\n      const r = yi - p;\n      ssr += r * r;\n      const t = yi - uY;\n      sst += t * t;\n    }\n  }\n  return sst === 0 ? 0 : 1 - ssr / sst;\n}\n\nexport function regressionLinear(\n  data: any[],\n  x: (d: any) => number = d => d.x,\n  y: (d: any) => number = d => d.y,\n  options?: {\n    alpha?: number;\n  }\n) {\n  const alpha = options?.alpha ?? 0.05;\n  const points = regressionPoints(data, x, y);\n  const model = fitRegression(points, 1);\n  const coef = regressionCoefficients(model);\n  const a = coef[0];\n  const b = coef[1];\n  const predict = (xx: number) => regressionPrediction(model, xx);\n  const comps = computeLinearCIComponents(data, x, y, predict);\n  let meanY = 0;\n  points.forEach((p, i) => {\n    meanY += (p.y - meanY) / (i + 1);\n  });\n\n  function evaluateGrid(N: number) {\n    return regressionGrid(comps.min, comps.max, N, predict);\n  }\n\n  function confidenceInterval(N: number = 50) {\n    const out: { x: number; mean: number; lower: number; upper: number; predLower: number; predUpper: number }[] = [];\n    if (comps.n === 0 || N <= 0) {\n      return out;\n    }\n    const z = confidenceCriticalValue(alpha);\n    if (comps.min === comps.max) {\n      const m = predict(comps.min);\n      const errs = stdErrorsAt(comps.min, comps);\n      const meanHalfWidth = confidenceHalfWidth(z, errs.seMean);\n      const predHalfWidth = confidenceHalfWidth(z, errs.sePred);\n      for (let i = 0; i < N; i++) {\n        out.push({\n          x: comps.min,\n          mean: m,\n          lower: m - meanHalfWidth,\n          upper: m + meanHalfWidth,\n          predLower: m - predHalfWidth,\n          predUpper: m + predHalfWidth\n        });\n      }\n      return out;\n    }\n    const step = (comps.max - comps.min) / (N - 1);\n    for (let i = 0; i < N; i++) {\n      const px = i === N - 1 ? comps.max : comps.min + step * i;\n      const m = predict(px);\n      const errs = stdErrorsAt(px, comps);\n      const meanHalfWidth = confidenceHalfWidth(z, errs.seMean);\n      const predHalfWidth = confidenceHalfWidth(z, errs.sePred);\n      out.push({\n        x: px,\n        mean: m,\n        lower: m - meanHalfWidth,\n        upper: m + meanHalfWidth,\n        predLower: m - predHalfWidth,\n        predUpper: m + predHalfWidth\n      });\n    }\n    return out;\n  }\n\n  return {\n    coef: { a, b },\n    rank: model.rank,\n    predict,\n    rSquared: rSquared(data, x, y, meanY, predict),\n    evaluateGrid,\n    confidenceInterval\n  };\n}\n\nexport default {\n  ordinaryLeastSquares,\n  visitPoints,\n  rSquared,\n  regressionLinear\n};\n"]}

package/es/common/regression-logistic.d.ts

@@ -1,3 +1,9 @@
+/**
+ * Simple logistic regression (binary) using Newton-Raphson (IRLS).
+ * Returns { coef, predict, evaluateGrid }
+ * - coef: [intercept, beta]
+ * - predict(x): probability p(y=1|x)
+ */
 export declare function regressionLogistic(data: any[], x?: (d: any) => number, y?: (d: any) => number, options?: {
     maxIteration?: number;
     tol?: number;

package/es/common/regression-logistic.js

@@ -1,8 +1,13 @@
 import { visitPoints } from "./regression-linear";
 
-import { computeLinearCIComponents, invNorm, stdErrorsAt } from "./regression-utils";
+import { computeLinearCIComponents, confidenceCriticalValue, confidenceHalfWidth, stdErrorsAt } from "./regression-utils";
 
-export function regressionLogistic(data, x = (d => d.x), y = (d => d.y), options) {
+/**
+ * Simple logistic regression (binary) using Newton-Raphson (IRLS).
+ * Returns { coef, predict, evaluateGrid }
+ * - coef: [intercept, beta]
+ * - predict(x): probability p(y=1|x)
+ */ export function regressionLogistic(data, x = (d => d.x), y = (d => d.y), options) {
     var _a, _b, _c;
     const maxIter = null !== (_a = null == options ? void 0 : options.maxIteration) && void 0 !== _a ? _a : 25, tol = null !== (_b = null == options ? void 0 : options.tol) && void 0 !== _b ? _b : 1e-6, alpha = null !== (_c = null == options ? void 0 : options.alpha) && void 0 !== _c ? _c : .05, xs = [], ys = [];
     visitPoints(data, x, y, ((dx, dy) => {
@@ -15,7 +20,8 @@ export function regressionLogistic(data, x = (d => d.x), y = (d => d.y), options
         evaluateGrid: N => [],
         confidenceInterval: (N = 50) => []
     };
-    let intercept = 0, beta = 0;
+    // initial coef via linear regression rough guess
+        let intercept = 0, beta = 0;
     for (let iter = 0; iter < maxIter; iter++) {
         const p = new Array(n);
         let converged = !0;
@@ -23,12 +29,14 @@ export function regressionLogistic(data, x = (d => d.x), y = (d => d.y), options
             const z = intercept + beta * xs[i], pi = 1 / (1 + Math.exp(-z));
             p[i] = pi;
         }
-        let g0 = 0, g1 = 0, h00 = 0, h01 = 0, h11 = 0;
+        // compute gradient and hessian
+                let g0 = 0, g1 = 0, h00 = 0, h01 = 0, h11 = 0;
         for (let i = 0; i < n; i++) {
             const wi = p[i] * (1 - p[i]), diff = ys[i] - p[i];
             g0 += diff, g1 += diff * xs[i], h00 += wi, h01 += wi * xs[i], h11 += wi * xs[i] * xs[i];
         }
-        const det = h00 * h11 - h01 * h01;
+        // solve 2x2 system H * delta = g
+                const det = h00 * h11 - h01 * h01;
         if (Math.abs(det) < 1e-12) break;
         const delta0 = (h11 * g0 - h01 * g1) / det, delta1 = (-h01 * g0 + h00 * g1) / det;
         if (intercept += delta0, beta += delta1, (Math.abs(delta0) > tol || Math.abs(delta1) > tol) && (converged = !1), 
@@ -71,29 +79,29 @@ export function regressionLogistic(data, x = (d => d.x), y = (d => d.y), options
             if (N <= 0) return out;
             const comps = computeLinearCIComponents(data, x, y, predict);
             if (0 === comps.n) return out;
-            const z = Math.abs(invNorm(1 - alpha / 2));
+            const z = confidenceCriticalValue(alpha);
             if (comps.min === comps.max) {
-                const v = predict(comps.min), errs = stdErrorsAt(comps.min, comps);
+                const v = predict(comps.min), errs = stdErrorsAt(comps.min, comps), meanHalfWidth = confidenceHalfWidth(z, errs.seMean), predHalfWidth = confidenceHalfWidth(z, errs.sePred);
                 for (let i = 0; i < N; i++) out.push({
                     x: comps.min,
                     mean: v,
-                    lower: v - z * errs.seMean,
-                    upper: v + z * errs.seMean,
-                    predLower: v - z * errs.sePred,
-                    predUpper: v + z * errs.sePred
+                    lower: v - meanHalfWidth,
+                    upper: v + meanHalfWidth,
+                    predLower: v - predHalfWidth,
+                    predUpper: v + predHalfWidth
                 });
                 return out;
             }
             const step = (comps.max - comps.min) / (N - 1);
             for (let i = 0; i < N; i++) {
-                const px = i === N - 1 ? comps.max : comps.min + step * i, yh = predict(px), errs = stdErrorsAt(px, comps);
+                const px = i === N - 1 ? comps.max : comps.min + step * i, yh = predict(px), errs = stdErrorsAt(px, comps), meanHalfWidth = confidenceHalfWidth(z, errs.seMean), predHalfWidth = confidenceHalfWidth(z, errs.sePred);
                 out.push({
                     x: px,
                     mean: yh,
-                    lower: yh - z * errs.seMean,
-                    upper: yh + z * errs.seMean,
-                    predLower: yh - z * errs.sePred,
-                    predUpper: yh + z * errs.sePred
+                    lower: yh - meanHalfWidth,
+                    upper: yh + meanHalfWidth,
+                    predLower: yh - predHalfWidth,
+                    predUpper: yh + predHalfWidth
                 });
             }
             return out;

package/es/common/regression-logistic.js.map

@@ -1 +1 @@
-{"version":3,"sources":["../src/common/regression-logistic.ts"],"names":[],"mappings":"AAAA,OAAO,EAAE,WAAW,EAAE,MAAM,qBAAqB,CAAC;AAClD,OAAO,EAAE,yBAAyB,EAAE,OAAO,EAAE,WAAW,EAAE,MAAM,oBAAoB,CAAC;AAQrF,MAAM,UAAU,kBAAkB,CAChC,IAAW,EACX,IAAwB,CAAC,CAAC,EAAE,CAAC,CAAC,CAAC,CAAC,EAChC,IAAwB,CAAC,CAAC,EAAE,CAAC,CAAC,CAAC,CAAC,EAChC,OAAiE;;IAEjE,MAAM,OAAO,GAAG,MAAA,OAAO,aAAP,OAAO,uBAAP,OAAO,CAAE,YAAY,mCAAI,EAAE,CAAC;IAC5C,MAAM,GAAG,GAAG,MAAA,OAAO,aAAP,OAAO,uBAAP,OAAO,CAAE,GAAG,mCAAI,IAAI,CAAC;IACjC,MAAM,KAAK,GAAG,MAAA,OAAO,aAAP,OAAO,uBAAP,OAAO,CAAE,KAAK,mCAAI,IAAI,CAAC;IAErC,MAAM,EAAE,GAAa,EAAE,CAAC;IACxB,MAAM,EAAE,GAAa,EAAE,CAAC;IACxB,WAAW,CAAC,IAAI,EAAE,CAAC,EAAE,CAAC,EAAE,CAAC,EAAE,EAAE,EAAE,EAAE,EAAE;QACjC,EAAE,CAAC,IAAI,CAAC,EAAE,CAAC,CAAC;QACZ,EAAE,CAAC,IAAI,CAAC,EAAE,CAAC,CAAC,CAAC,CAAC,CAAC,CAAC,CAAC,CAAC,CAAC,CAAC;IACtB,CAAC,CAAC,CAAC;IAEH,MAAM,CAAC,GAAG,EAAE,CAAC,MAAM,CAAC;IACpB,IAAI,CAAC,KAAK,CAAC,EAAE;QACX,OAAO;YACL,IAAI,EAAE,CAAC,CAAC,EAAE,CAAC,CAAC;YACZ,OAAO,EAAE,CAAC,EAAU,EAAE,EAAE,CAAC,CAAC;YAC1B,YAAY,EAAE,CAAC,CAAS,EAAE,EAAE,CAAC,EAAgC;YAC7D,kBAAkB,EAAE,CAAC,IAAY,EAAE,EAAE,EAAE,CACrC,EAAuG;SAC1G,CAAC;KACH;IAGD,IAAI,SAAS,GAAG,CAAC,CAAC;IAClB,IAAI,IAAI,GAAG,CAAC,CAAC;IAEb,KAAK,IAAI,IAAI,GAAG,CAAC,EAAE,IAAI,GAAG,OAAO,EAAE,IAAI,EAAE,EAAE;QACzC,MAAM,CAAC,GAAa,IAAI,KAAK,CAAC,CAAC,CAAC,CAAC;QACjC,IAAI,SAAS,GAAG,IAAI,CAAC;QACrB,KAAK,IAAI,CAAC,GAAG,CAAC,EAAE,CAAC,GAAG,CAAC,EAAE,CAAC,EAAE,EAAE;YAC1B,MAAM,CAAC,GAAG,SAAS,GAAG,IAAI,GAAG,EAAE,CAAC,CAAC,CAAC,CAAC;YACnC,MAAM,EAAE,GAAG,CAAC,GAAG,CAAC,CAAC,GAAG,IAAI,CAAC,GAAG,CAAC,CAAC,CAAC,CAAC,CAAC,CAAC;YAClC,CAAC,CAAC,CAAC,CAAC,GAAG,EAAE,CAAC;SACX;QAGD,IAAI,EAAE,GAAG,CAAC,CAAC;QACX,IAAI,EAAE,GAAG,CAAC,CAAC;QACX,IAAI,GAAG,GAAG,CAAC,CAAC;QACZ,IAAI,GAAG,GAAG,CAAC,CAAC;QACZ,IAAI,GAAG,GAAG,CAAC,CAAC;QACZ,KAAK,IAAI,CAAC,GAAG,CAAC,EAAE,CAAC,GAAG,CAAC,EAAE,CAAC,EAAE,EAAE;YAC1B,MAAM,EAAE,GAAG,CAAC,CAAC,CAAC,CAAC,GAAG,CAAC,CAAC,GAAG,CAAC,CAAC,CAAC,CAAC,CAAC,CAAC;YAC7B,MAAM,IAAI,GAAG,EAAE,CAAC,CAAC,CAAC,GAAG,CAAC,CAAC,CAAC,CAAC,CAAC;YAC1B,EAAE,IAAI,IAAI,CAAC;YACX,EAAE,IAAI,IAAI,GAAG,EAAE,CAAC,CAAC,CAAC,CAAC;YACnB,GAAG,IAAI,EAAE,CAAC;YACV,GAAG,IAAI,EAAE,GAAG,EAAE,CAAC,CAAC,CAAC,CAAC;YAClB,GAAG,IAAI,EAAE,GAAG,EAAE,CAAC,CAAC,CAAC,GAAG,EAAE,CAAC,CAAC,CAAC,CAAC;SAC3B;QAGD,MAAM,GAAG,GAAG,GAAG,GAAG,GAAG,GAAG,GAAG,GAAG,GAAG,CAAC;QAClC,IAAI,IAAI,CAAC,GAAG,CAAC,GAAG,CAAC,GAAG,KAAK,EAAE;YACzB,MAAM;SACP;QACD,MAAM,MAAM,GAAG,CAAC,GAAG,GAAG,EAAE,GAAG,GAAG,GAAG,EAAE,CAAC,GAAG,GAAG,CAAC;QAC3C,MAAM,MAAM,GAAG,CAAC,CAAC,GAAG,GAAG,EAAE,GAAG,GAAG,GAAG,EAAE,CAAC,GAAG,GAAG,CAAC;QAE5C,SAAS,IAAI,MAAM,CAAC;QACpB,IAAI,IAAI,MAAM,CAAC;QAEf,IAAI,IAAI,CAAC,GAAG,CAAC,MAAM,CAAC,GAAG,GAAG,IAAI,IAAI,CAAC,GAAG,CAAC,MAAM,CAAC,GAAG,GAAG,EAAE;YACpD,SAAS,GAAG,KAAK,CAAC;SACnB;QACD,IAAI,SAAS,EAAE;YACb,MAAM;SACP;KACF;IAED,MAAM,OAAO,GAAG,CAAC,EAAU,EAAE,EAAE;QAC7B,MAAM,CAAC,GAAG,SAAS,GAAG,IAAI,GAAG,EAAE,CAAC;QAChC,OAAO,CAAC,GAAG,CAAC,CAAC,GAAG,IAAI,CAAC,GAAG,CAAC,CAAC,CAAC,CAAC,CAAC,CAAC;IAChC,CAAC,CAAC;IAEF,SAAS,YAAY,CAAC,CAAS;QAC7B,MAAM,GAAG,GAA+B,EAAE,CAAC;QAC3C,IAAI,CAAC,IAAI,CAAC,EAAE;YACV,OAAO,GAAG,CAAC;SACZ;QACD,IAAI,GAAG,GAAG,QAAQ,CAAC;QACnB,IAAI,GAAG,GAAG,CAAC,QAAQ,CAAC;QACpB,WAAW,CAAC,IAAI,EAAE,CAAC,EAAE,CAAC,EAAE,EAAE,CAAC,EAAE;YAC3B,IAAI,EAAE,GAAG,GAAG,EAAE;gBACZ,GAAG,GAAG,EAAE,CAAC;aACV;YACD,IAAI,EAAE,GAAG,GAAG,EAAE;gBACZ,GAAG,GAAG,EAAE,CAAC;aACV;QACH,CAAC,CAAC,CAAC;QACH,IAAI,GAAG,KAAK,QAAQ,IAAI,GAAG,KAAK,CAAC,QAAQ,EAAE;YACzC,OAAO,GAAG,CAAC;SACZ;QACD,IAAI,GAAG,KAAK,GAAG,EAAE;YACf,MAAM,CAAC,GAAG,OAAO,CAAC,GAAG,CAAC,CAAC;YACvB,KAAK,IAAI,CAAC,GAAG,CAAC,EAAE,CAAC,GAAG,CAAC,EAAE,CAAC,EAAE,EAAE;gBAC1B,GAAG,CAAC,IAAI,CAAC,EAAE,CAAC,EAAE,GAAG,EAAE,CAAC,EAAE,CAAC,EAAE,CAAC,CAAC;aAC5B;YACD,OAAO,GAAG,CAAC;SACZ;QACD,MAAM,IAAI,GAAG,CAAC,GAAG,GAAG,GAAG,CAAC,GAAG,CAAC,CAAC,GAAG,CAAC,CAAC,CAAC;QACnC,KAAK,IAAI,CAAC,GAAG,CAAC,EAAE,CAAC,GAAG,CAAC,EAAE,CAAC,EAAE,EAAE;YAC1B,MAAM,EAAE,GAAG,CAAC,KAAK,CAAC,GAAG,CAAC,CAAC,CAAC,CAAC,GAAG,CAAC,CAAC,CAAC,GAAG,GAAG,IAAI,GAAG,CAAC,CAAC;YAC9C,GAAG,CAAC,IAAI,CAAC,EAAE,CAAC,EAAE,EAAE,EAAE,CAAC,EAAE,OAAO,CAAC,EAAE,CAAC,EAAE,CAAC,CAAC;SACrC;QACD,OAAO,GAAG,CAAC;IACb,CAAC;IAED,SAAS,kBAAkB,CAAC,IAAY,EAAE;QACxC,MAAM,GAAG,GAAsG,EAAE,CAAC;QAElH,IAAI,CAAC,IAAI,CAAC,EAAE;YACV,OAAO,GAAG,CAAC;SACZ;QAED,MAAM,KAAK,GAAG,yBAAyB,CAAC,IAAI,EAAE,CAAC,EAAE,CAAC,EAAE,OAAO,CAAC,CAAC;QAC7D,IAAI,KAAK,CAAC,CAAC,KAAK,CAAC,EAAE;YACjB,OAAO,GAAG,CAAC;SACZ;QAED,MAAM,CAAC,GAAG,IAAI,CAAC,GAAG,CAAC,OAAO,CAAC,CAAC,GAAG,KAAK,GAAG,CAAC,CAAC,CAAC,CAAC;QAC3C,IAAI,KAAK,CAAC,GAAG,KAAK,KAAK,CAAC,GAAG,EAAE;YAC3B,MAAM,CAAC,GAAG,OAAO,CAAC,KAAK,CAAC,GAAG,CAAC,CAAC;YAC7B,MAAM,IAAI,GAAG,WAAW,CAAC,KAAK,CAAC,GAAG,EAAE,KAAK,CAAC,CAAC;YAC3C,KAAK,IAAI,CAAC,GAAG,CAAC,EAAE,CAAC,GAAG,CAAC,EAAE,CAAC,EAAE,EAAE;gBAC1B,GAAG,CAAC,IAAI,CAAC;oBACP,CAAC,EAAE,KAAK,CAAC,GAAG;oBACZ,IAAI,EAAE,CAAC;oBACP,KAAK,EAAE,CAAC,GAAG,CAAC,GAAG,IAAI,CAAC,MAAM;oBAC1B,KAAK,EAAE,CAAC,GAAG,CAAC,GAAG,IAAI,CAAC,MAAM;oBAC1B,SAAS,EAAE,CAAC,GAAG,CAAC,GAAG,IAAI,CAAC,MAAM;oBAC9B,SAAS,EAAE,CAAC,GAAG,CAAC,GAAG,IAAI,CAAC,MAAM;iBAC/B,CAAC,CAAC;aACJ;YACD,OAAO,GAAG,CAAC;SACZ;QAED,MAAM,IAAI,GAAG,CAAC,KAAK,CAAC,GAAG,GAAG,KAAK,CAAC,GAAG,CAAC,GAAG,CAAC,CAAC,GAAG,CAAC,CAAC,CAAC;QAC/C,KAAK,IAAI,CAAC,GAAG,CAAC,EAAE,CAAC,GAAG,CAAC,EAAE,CAAC,EAAE,EAAE;YAC1B,MAAM,EAAE,GAAG,CAAC,KAAK,CAAC,GAAG,CAAC,CAAC,CAAC,CAAC,KAAK,CAAC,GAAG,CAAC,CAAC,CAAC,KAAK,CAAC,GAAG,GAAG,IAAI,GAAG,CAAC,CAAC;YAC1D,MAAM,EAAE,GAAG,OAAO,CAAC,EAAE,CAAC,CAAC;YACvB,MAAM,IAAI,GAAG,WAAW,CAAC,EAAE,EAAE,KAAK,CAAC,CAAC;YACpC,GAAG,CAAC,IAAI,CAAC;gBACP,CAAC,EAAE,EAAE;gBACL,IAAI,EAAE,EAAE;gBACR,KAAK,EAAE,EAAE,GAAG,CAAC,GAAG,IAAI,CAAC,MAAM;gBAC3B,KAAK,EAAE,EAAE,GAAG,CAAC,GAAG,IAAI,CAAC,MAAM;gBAC3B,SAAS,EAAE,EAAE,GAAG,CAAC,GAAG,IAAI,CAAC,MAAM;gBAC/B,SAAS,EAAE,EAAE,GAAG,CAAC,GAAG,IAAI,CAAC,MAAM;aAChC,CAAC,CAAC;SACJ;QACD,OAAO,GAAG,CAAC;IACb,CAAC;IAED,OAAO;QACL,IAAI,EAAE,CAAC,SAAS,EAAE,IAAI,CAAC;QACvB,OAAO;QACP,YAAY;QACZ,kBAAkB;KACnB,CAAC;AACJ,CAAC;AAED,eAAe,kBAAkB,CAAC","file":"regression-logistic.js","sourcesContent":["import { visitPoints } from './regression-linear';\nimport { computeLinearCIComponents, invNorm, stdErrorsAt } from './regression-utils';\n\n/**\n * Simple logistic regression (binary) using Newton-Raphson (IRLS).\n * Returns { coef, predict, evaluateGrid }\n * - coef: [intercept, beta]\n * - predict(x): probability p(y=1|x)\n */\nexport function regressionLogistic(\n  data: any[],\n  x: (d: any) => number = d => d.x,\n  y: (d: any) => number = d => d.y,\n  options?: { maxIteration?: number; tol?: number; alpha?: number }\n) {\n  const maxIter = options?.maxIteration ?? 25;\n  const tol = options?.tol ?? 1e-6;\n  const alpha = options?.alpha ?? 0.05;\n  // build arrays\n  const xs: number[] = [];\n  const ys: number[] = [];\n  visitPoints(data, x, y, (dx, dy) => {\n    xs.push(dx);\n    ys.push(dy ? 1 : 0);\n  });\n\n  const n = xs.length;\n  if (n === 0) {\n    return {\n      coef: [0, 0],\n      predict: (_x: number) => 0,\n      evaluateGrid: (N: number) => [] as { x: number; y: number }[],\n      confidenceInterval: (N: number = 50) =>\n        [] as { x: number; mean: number; lower: number; upper: number; predLower: number; predUpper: number }[]\n    };\n  }\n\n  // initial coef via linear regression rough guess\n  let intercept = 0;\n  let beta = 0;\n\n  for (let iter = 0; iter < maxIter; iter++) {\n    const p: number[] = new Array(n);\n    let converged = true;\n    for (let i = 0; i < n; i++) {\n      const z = intercept + beta * xs[i];\n      const pi = 1 / (1 + Math.exp(-z));\n      p[i] = pi;\n    }\n\n    // compute gradient and hessian\n    let g0 = 0;\n    let g1 = 0;\n    let h00 = 0;\n    let h01 = 0;\n    let h11 = 0;\n    for (let i = 0; i < n; i++) {\n      const wi = p[i] * (1 - p[i]);\n      const diff = ys[i] - p[i];\n      g0 += diff;\n      g1 += diff * xs[i];\n      h00 += wi;\n      h01 += wi * xs[i];\n      h11 += wi * xs[i] * xs[i];\n    }\n\n    // solve 2x2 system H * delta = g\n    const det = h00 * h11 - h01 * h01;\n    if (Math.abs(det) < 1e-12) {\n      break;\n    }\n    const delta0 = (h11 * g0 - h01 * g1) / det;\n    const delta1 = (-h01 * g0 + h00 * g1) / det;\n\n    intercept += delta0;\n    beta += delta1;\n\n    if (Math.abs(delta0) > tol || Math.abs(delta1) > tol) {\n      converged = false;\n    }\n    if (converged) {\n      break;\n    }\n  }\n\n  const predict = (xx: number) => {\n    const z = intercept + beta * xx;\n    return 1 / (1 + Math.exp(-z));\n  };\n\n  function evaluateGrid(N: number) {\n    const out: { x: number; y: number }[] = [];\n    if (N <= 0) {\n      return out;\n    }\n    let min = Infinity;\n    let max = -Infinity;\n    visitPoints(data, x, y, dx => {\n      if (dx < min) {\n        min = dx;\n      }\n      if (dx > max) {\n        max = dx;\n      }\n    });\n    if (min === Infinity || max === -Infinity) {\n      return out;\n    }\n    if (min === max) {\n      const v = predict(min);\n      for (let i = 0; i < N; i++) {\n        out.push({ x: min, y: v });\n      }\n      return out;\n    }\n    const step = (max - min) / (N - 1);\n    for (let i = 0; i < N; i++) {\n      const px = i === N - 1 ? max : min + step * i;\n      out.push({ x: px, y: predict(px) });\n    }\n    return out;\n  }\n\n  function confidenceInterval(N: number = 50) {\n    const out: { x: number; mean: number; lower: number; upper: number; predLower: number; predUpper: number }[] = [];\n\n    if (N <= 0) {\n      return out;\n    }\n\n    const comps = computeLinearCIComponents(data, x, y, predict);\n    if (comps.n === 0) {\n      return out;\n    }\n\n    const z = Math.abs(invNorm(1 - alpha / 2));\n    if (comps.min === comps.max) {\n      const v = predict(comps.min);\n      const errs = stdErrorsAt(comps.min, comps);\n      for (let i = 0; i < N; i++) {\n        out.push({\n          x: comps.min,\n          mean: v,\n          lower: v - z * errs.seMean,\n          upper: v + z * errs.seMean,\n          predLower: v - z * errs.sePred,\n          predUpper: v + z * errs.sePred\n        });\n      }\n      return out;\n    }\n\n    const step = (comps.max - comps.min) / (N - 1);\n    for (let i = 0; i < N; i++) {\n      const px = i === N - 1 ? comps.max : comps.min + step * i;\n      const yh = predict(px);\n      const errs = stdErrorsAt(px, comps);\n      out.push({\n        x: px,\n        mean: yh,\n        lower: yh - z * errs.seMean,\n        upper: yh + z * errs.seMean,\n        predLower: yh - z * errs.sePred,\n        predUpper: yh + z * errs.sePred\n      });\n    }\n    return out;\n  }\n\n  return {\n    coef: [intercept, beta],\n    predict,\n    evaluateGrid,\n    confidenceInterval\n  };\n}\n\nexport default regressionLogistic;\n"]}
+{"version":3,"sources":["../src/common/regression-logistic.ts"],"names":[],"mappings":"AAAA,OAAO,EAAE,WAAW,EAAE,MAAM,qBAAqB,CAAC;AAClD,OAAO,EACL,yBAAyB,EACzB,uBAAuB,EACvB,mBAAmB,EACnB,WAAW,EACZ,MAAM,oBAAoB,CAAC;AAE5B;;;;;GAKG;AACH,MAAM,UAAU,kBAAkB,CAChC,IAAW,EACX,IAAwB,CAAC,CAAC,EAAE,CAAC,CAAC,CAAC,CAAC,EAChC,IAAwB,CAAC,CAAC,EAAE,CAAC,CAAC,CAAC,CAAC,EAChC,OAAiE;;IAEjE,MAAM,OAAO,GAAG,MAAA,OAAO,aAAP,OAAO,uBAAP,OAAO,CAAE,YAAY,mCAAI,EAAE,CAAC;IAC5C,MAAM,GAAG,GAAG,MAAA,OAAO,aAAP,OAAO,uBAAP,OAAO,CAAE,GAAG,mCAAI,IAAI,CAAC;IACjC,MAAM,KAAK,GAAG,MAAA,OAAO,aAAP,OAAO,uBAAP,OAAO,CAAE,KAAK,mCAAI,IAAI,CAAC;IACrC,eAAe;IACf,MAAM,EAAE,GAAa,EAAE,CAAC;IACxB,MAAM,EAAE,GAAa,EAAE,CAAC;IACxB,WAAW,CAAC,IAAI,EAAE,CAAC,EAAE,CAAC,EAAE,CAAC,EAAE,EAAE,EAAE,EAAE,EAAE;QACjC,EAAE,CAAC,IAAI,CAAC,EAAE,CAAC,CAAC;QACZ,EAAE,CAAC,IAAI,CAAC,EAAE,CAAC,CAAC,CAAC,CAAC,CAAC,CAAC,CAAC,CAAC,CAAC,CAAC;IACtB,CAAC,CAAC,CAAC;IAEH,MAAM,CAAC,GAAG,EAAE,CAAC,MAAM,CAAC;IACpB,IAAI,CAAC,KAAK,CAAC,EAAE;QACX,OAAO;YACL,IAAI,EAAE,CAAC,CAAC,EAAE,CAAC,CAAC;YACZ,OAAO,EAAE,CAAC,EAAU,EAAE,EAAE,CAAC,CAAC;YAC1B,YAAY,EAAE,CAAC,CAAS,EAAE,EAAE,CAAC,EAAgC;YAC7D,kBAAkB,EAAE,CAAC,IAAY,EAAE,EAAE,EAAE,CACrC,EAAuG;SAC1G,CAAC;KACH;IAED,iDAAiD;IACjD,IAAI,SAAS,GAAG,CAAC,CAAC;IAClB,IAAI,IAAI,GAAG,CAAC,CAAC;IAEb,KAAK,IAAI,IAAI,GAAG,CAAC,EAAE,IAAI,GAAG,OAAO,EAAE,IAAI,EAAE,EAAE;QACzC,MAAM,CAAC,GAAa,IAAI,KAAK,CAAC,CAAC,CAAC,CAAC;QACjC,IAAI,SAAS,GAAG,IAAI,CAAC;QACrB,KAAK,IAAI,CAAC,GAAG,CAAC,EAAE,CAAC,GAAG,CAAC,EAAE,CAAC,EAAE,EAAE;YAC1B,MAAM,CAAC,GAAG,SAAS,GAAG,IAAI,GAAG,EAAE,CAAC,CAAC,CAAC,CAAC;YACnC,MAAM,EAAE,GAAG,CAAC,GAAG,CAAC,CAAC,GAAG,IAAI,CAAC,GAAG,CAAC,CAAC,CAAC,CAAC,CAAC,CAAC;YAClC,CAAC,CAAC,CAAC,CAAC,GAAG,EAAE,CAAC;SACX;QAED,+BAA+B;QAC/B,IAAI,EAAE,GAAG,CAAC,CAAC;QACX,IAAI,EAAE,GAAG,CAAC,CAAC;QACX,IAAI,GAAG,GAAG,CAAC,CAAC;QACZ,IAAI,GAAG,GAAG,CAAC,CAAC;QACZ,IAAI,GAAG,GAAG,CAAC,CAAC;QACZ,KAAK,IAAI,CAAC,GAAG,CAAC,EAAE,CAAC,GAAG,CAAC,EAAE,CAAC,EAAE,EAAE;YAC1B,MAAM,EAAE,GAAG,CAAC,CAAC,CAAC,CAAC,GAAG,CAAC,CAAC,GAAG,CAAC,CAAC,CAAC,CAAC,CAAC,CAAC;YAC7B,MAAM,IAAI,GAAG,EAAE,CAAC,CAAC,CAAC,GAAG,CAAC,CAAC,CAAC,CAAC,CAAC;YAC1B,EAAE,IAAI,IAAI,CAAC;YACX,EAAE,IAAI,IAAI,GAAG,EAAE,CAAC,CAAC,CAAC,CAAC;YACnB,GAAG,IAAI,EAAE,CAAC;YACV,GAAG,IAAI,EAAE,GAAG,EAAE,CAAC,CAAC,CAAC,CAAC;YAClB,GAAG,IAAI,EAAE,GAAG,EAAE,CAAC,CAAC,CAAC,GAAG,EAAE,CAAC,CAAC,CAAC,CAAC;SAC3B;QAED,iCAAiC;QACjC,MAAM,GAAG,GAAG,GAAG,GAAG,GAAG,GAAG,GAAG,GAAG,GAAG,CAAC;QAClC,IAAI,IAAI,CAAC,GAAG,CAAC,GAAG,CAAC,GAAG,KAAK,EAAE;YACzB,MAAM;SACP;QACD,MAAM,MAAM,GAAG,CAAC,GAAG,GAAG,EAAE,GAAG,GAAG,GAAG,EAAE,CAAC,GAAG,GAAG,CAAC;QAC3C,MAAM,MAAM,GAAG,CAAC,CAAC,GAAG,GAAG,EAAE,GAAG,GAAG,GAAG,EAAE,CAAC,GAAG,GAAG,CAAC;QAE5C,SAAS,IAAI,MAAM,CAAC;QACpB,IAAI,IAAI,MAAM,CAAC;QAEf,IAAI,IAAI,CAAC,GAAG,CAAC,MAAM,CAAC,GAAG,GAAG,IAAI,IAAI,CAAC,GAAG,CAAC,MAAM,CAAC,GAAG,GAAG,EAAE;YACpD,SAAS,GAAG,KAAK,CAAC;SACnB;QACD,IAAI,SAAS,EAAE;YACb,MAAM;SACP;KACF;IAED,MAAM,OAAO,GAAG,CAAC,EAAU,EAAE,EAAE;QAC7B,MAAM,CAAC,GAAG,SAAS,GAAG,IAAI,GAAG,EAAE,CAAC;QAChC,OAAO,CAAC,GAAG,CAAC,CAAC,GAAG,IAAI,CAAC,GAAG,CAAC,CAAC,CAAC,CAAC,CAAC,CAAC;IAChC,CAAC,CAAC;IAEF,SAAS,YAAY,CAAC,CAAS;QAC7B,MAAM,GAAG,GAA+B,EAAE,CAAC;QAC3C,IAAI,CAAC,IAAI,CAAC,EAAE;YACV,OAAO,GAAG,CAAC;SACZ;QACD,IAAI,GAAG,GAAG,QAAQ,CAAC;QACnB,IAAI,GAAG,GAAG,CAAC,QAAQ,CAAC;QACpB,WAAW,CAAC,IAAI,EAAE,CAAC,EAAE,CAAC,EAAE,EAAE,CAAC,EAAE;YAC3B,IAAI,EAAE,GAAG,GAAG,EAAE;gBACZ,GAAG,GAAG,EAAE,CAAC;aACV;YACD,IAAI,EAAE,GAAG,GAAG,EAAE;gBACZ,GAAG,GAAG,EAAE,CAAC;aACV;QACH,CAAC,CAAC,CAAC;QACH,IAAI,GAAG,KAAK,QAAQ,IAAI,GAAG,KAAK,CAAC,QAAQ,EAAE;YACzC,OAAO,GAAG,CAAC;SACZ;QACD,IAAI,GAAG,KAAK,GAAG,EAAE;YACf,MAAM,CAAC,GAAG,OAAO,CAAC,GAAG,CAAC,CAAC;YACvB,KAAK,IAAI,CAAC,GAAG,CAAC,EAAE,CAAC,GAAG,CAAC,EAAE,CAAC,EAAE,EAAE;gBAC1B,GAAG,CAAC,IAAI,CAAC,EAAE,CAAC,EAAE,GAAG,EAAE,CAAC,EAAE,CAAC,EAAE,CAAC,CAAC;aAC5B;YACD,OAAO,GAAG,CAAC;SACZ;QACD,MAAM,IAAI,GAAG,CAAC,GAAG,GAAG,GAAG,CAAC,GAAG,CAAC,CAAC,GAAG,CAAC,CAAC,CAAC;QACnC,KAAK,IAAI,CAAC,GAAG,CAAC,EAAE,CAAC,GAAG,CAAC,EAAE,CAAC,EAAE,EAAE;YAC1B,MAAM,EAAE,GAAG,CAAC,KAAK,CAAC,GAAG,CAAC,CAAC,CAAC,CAAC,GAAG,CAAC,CAAC,CAAC,GAAG,GAAG,IAAI,GAAG,CAAC,CAAC;YAC9C,GAAG,CAAC,IAAI,CAAC,EAAE,CAAC,EAAE,EAAE,EAAE,CAAC,EAAE,OAAO,CAAC,EAAE,CAAC,EAAE,CAAC,CAAC;SACrC;QACD,OAAO,GAAG,CAAC;IACb,CAAC;IAED,SAAS,kBAAkB,CAAC,IAAY,EAAE;QACxC,MAAM,GAAG,GAAsG,EAAE,CAAC;QAElH,IAAI,CAAC,IAAI,CAAC,EAAE;YACV,OAAO,GAAG,CAAC;SACZ;QAED,MAAM,KAAK,GAAG,yBAAyB,CAAC,IAAI,EAAE,CAAC,EAAE,CAAC,EAAE,OAAO,CAAC,CAAC;QAC7D,IAAI,KAAK,CAAC,CAAC,KAAK,CAAC,EAAE;YACjB,OAAO,GAAG,CAAC;SACZ;QAED,MAAM,CAAC,GAAG,uBAAuB,CAAC,KAAK,CAAC,CAAC;QACzC,IAAI,KAAK,CAAC,GAAG,KAAK,KAAK,CAAC,GAAG,EAAE;YAC3B,MAAM,CAAC,GAAG,OAAO,CAAC,KAAK,CAAC,GAAG,CAAC,CAAC;YAC7B,MAAM,IAAI,GAAG,WAAW,CAAC,KAAK,CAAC,GAAG,EAAE,KAAK,CAAC,CAAC;YAC3C,MAAM,aAAa,GAAG,mBAAmB,CAAC,CAAC,EAAE,IAAI,CAAC,MAAM,CAAC,CAAC;YAC1D,MAAM,aAAa,GAAG,mBAAmB,CAAC,CAAC,EAAE,IAAI,CAAC,MAAM,CAAC,CAAC;YAC1D,KAAK,IAAI,CAAC,GAAG,CAAC,EAAE,CAAC,GAAG,CAAC,EAAE,CAAC,EAAE,EAAE;gBAC1B,GAAG,CAAC,IAAI,CAAC;oBACP,CAAC,EAAE,KAAK,CAAC,GAAG;oBACZ,IAAI,EAAE,CAAC;oBACP,KAAK,EAAE,CAAC,GAAG,aAAa;oBACxB,KAAK,EAAE,CAAC,GAAG,aAAa;oBACxB,SAAS,EAAE,CAAC,GAAG,aAAa;oBAC5B,SAAS,EAAE,CAAC,GAAG,aAAa;iBAC7B,CAAC,CAAC;aACJ;YACD,OAAO,GAAG,CAAC;SACZ;QAED,MAAM,IAAI,GAAG,CAAC,KAAK,CAAC,GAAG,GAAG,KAAK,CAAC,GAAG,CAAC,GAAG,CAAC,CAAC,GAAG,CAAC,CAAC,CAAC;QAC/C,KAAK,IAAI,CAAC,GAAG,CAAC,EAAE,CAAC,GAAG,CAAC,EAAE,CAAC,EAAE,EAAE;YAC1B,MAAM,EAAE,GAAG,CAAC,KAAK,CAAC,GAAG,CAAC,CAAC,CAAC,CAAC,KAAK,CAAC,GAAG,CAAC,CAAC,CAAC,KAAK,CAAC,GAAG,GAAG,IAAI,GAAG,CAAC,CAAC;YAC1D,MAAM,EAAE,GAAG,OAAO,CAAC,EAAE,CAAC,CAAC;YACvB,MAAM,IAAI,GAAG,WAAW,CAAC,EAAE,EAAE,KAAK,CAAC,CAAC;YACpC,MAAM,aAAa,GAAG,mBAAmB,CAAC,CAAC,EAAE,IAAI,CAAC,MAAM,CAAC,CAAC;YAC1D,MAAM,aAAa,GAAG,mBAAmB,CAAC,CAAC,EAAE,IAAI,CAAC,MAAM,CAAC,CAAC;YAC1D,GAAG,CAAC,IAAI,CAAC;gBACP,CAAC,EAAE,EAAE;gBACL,IAAI,EAAE,EAAE;gBACR,KAAK,EAAE,EAAE,GAAG,aAAa;gBACzB,KAAK,EAAE,EAAE,GAAG,aAAa;gBACzB,SAAS,EAAE,EAAE,GAAG,aAAa;gBAC7B,SAAS,EAAE,EAAE,GAAG,aAAa;aAC9B,CAAC,CAAC;SACJ;QACD,OAAO,GAAG,CAAC;IACb,CAAC;IAED,OAAO;QACL,IAAI,EAAE,CAAC,SAAS,EAAE,IAAI,CAAC;QACvB,OAAO;QACP,YAAY;QACZ,kBAAkB;KACnB,CAAC;AACJ,CAAC;AAED,eAAe,kBAAkB,CAAC","file":"regression-logistic.js","sourcesContent":["import { visitPoints } from './regression-linear';\nimport {\n  computeLinearCIComponents,\n  confidenceCriticalValue,\n  confidenceHalfWidth,\n  stdErrorsAt\n} from './regression-utils';\n\n/**\n * Simple logistic regression (binary) using Newton-Raphson (IRLS).\n * Returns { coef, predict, evaluateGrid }\n * - coef: [intercept, beta]\n * - predict(x): probability p(y=1|x)\n */\nexport function regressionLogistic(\n  data: any[],\n  x: (d: any) => number = d => d.x,\n  y: (d: any) => number = d => d.y,\n  options?: { maxIteration?: number; tol?: number; alpha?: number }\n) {\n  const maxIter = options?.maxIteration ?? 25;\n  const tol = options?.tol ?? 1e-6;\n  const alpha = options?.alpha ?? 0.05;\n  // build arrays\n  const xs: number[] = [];\n  const ys: number[] = [];\n  visitPoints(data, x, y, (dx, dy) => {\n    xs.push(dx);\n    ys.push(dy ? 1 : 0);\n  });\n\n  const n = xs.length;\n  if (n === 0) {\n    return {\n      coef: [0, 0],\n      predict: (_x: number) => 0,\n      evaluateGrid: (N: number) => [] as { x: number; y: number }[],\n      confidenceInterval: (N: number = 50) =>\n        [] as { x: number; mean: number; lower: number; upper: number; predLower: number; predUpper: number }[]\n    };\n  }\n\n  // initial coef via linear regression rough guess\n  let intercept = 0;\n  let beta = 0;\n\n  for (let iter = 0; iter < maxIter; iter++) {\n    const p: number[] = new Array(n);\n    let converged = true;\n    for (let i = 0; i < n; i++) {\n      const z = intercept + beta * xs[i];\n      const pi = 1 / (1 + Math.exp(-z));\n      p[i] = pi;\n    }\n\n    // compute gradient and hessian\n    let g0 = 0;\n    let g1 = 0;\n    let h00 = 0;\n    let h01 = 0;\n    let h11 = 0;\n    for (let i = 0; i < n; i++) {\n      const wi = p[i] * (1 - p[i]);\n      const diff = ys[i] - p[i];\n      g0 += diff;\n      g1 += diff * xs[i];\n      h00 += wi;\n      h01 += wi * xs[i];\n      h11 += wi * xs[i] * xs[i];\n    }\n\n    // solve 2x2 system H * delta = g\n    const det = h00 * h11 - h01 * h01;\n    if (Math.abs(det) < 1e-12) {\n      break;\n    }\n    const delta0 = (h11 * g0 - h01 * g1) / det;\n    const delta1 = (-h01 * g0 + h00 * g1) / det;\n\n    intercept += delta0;\n    beta += delta1;\n\n    if (Math.abs(delta0) > tol || Math.abs(delta1) > tol) {\n      converged = false;\n    }\n    if (converged) {\n      break;\n    }\n  }\n\n  const predict = (xx: number) => {\n    const z = intercept + beta * xx;\n    return 1 / (1 + Math.exp(-z));\n  };\n\n  function evaluateGrid(N: number) {\n    const out: { x: number; y: number }[] = [];\n    if (N <= 0) {\n      return out;\n    }\n    let min = Infinity;\n    let max = -Infinity;\n    visitPoints(data, x, y, dx => {\n      if (dx < min) {\n        min = dx;\n      }\n      if (dx > max) {\n        max = dx;\n      }\n    });\n    if (min === Infinity || max === -Infinity) {\n      return out;\n    }\n    if (min === max) {\n      const v = predict(min);\n      for (let i = 0; i < N; i++) {\n        out.push({ x: min, y: v });\n      }\n      return out;\n    }\n    const step = (max - min) / (N - 1);\n    for (let i = 0; i < N; i++) {\n      const px = i === N - 1 ? max : min + step * i;\n      out.push({ x: px, y: predict(px) });\n    }\n    return out;\n  }\n\n  function confidenceInterval(N: number = 50) {\n    const out: { x: number; mean: number; lower: number; upper: number; predLower: number; predUpper: number }[] = [];\n\n    if (N <= 0) {\n      return out;\n    }\n\n    const comps = computeLinearCIComponents(data, x, y, predict);\n    if (comps.n === 0) {\n      return out;\n    }\n\n    const z = confidenceCriticalValue(alpha);\n    if (comps.min === comps.max) {\n      const v = predict(comps.min);\n      const errs = stdErrorsAt(comps.min, comps);\n      const meanHalfWidth = confidenceHalfWidth(z, errs.seMean);\n      const predHalfWidth = confidenceHalfWidth(z, errs.sePred);\n      for (let i = 0; i < N; i++) {\n        out.push({\n          x: comps.min,\n          mean: v,\n          lower: v - meanHalfWidth,\n          upper: v + meanHalfWidth,\n          predLower: v - predHalfWidth,\n          predUpper: v + predHalfWidth\n        });\n      }\n      return out;\n    }\n\n    const step = (comps.max - comps.min) / (N - 1);\n    for (let i = 0; i < N; i++) {\n      const px = i === N - 1 ? comps.max : comps.min + step * i;\n      const yh = predict(px);\n      const errs = stdErrorsAt(px, comps);\n      const meanHalfWidth = confidenceHalfWidth(z, errs.seMean);\n      const predHalfWidth = confidenceHalfWidth(z, errs.sePred);\n      out.push({\n        x: px,\n        mean: yh,\n        lower: yh - meanHalfWidth,\n        upper: yh + meanHalfWidth,\n        predLower: yh - predHalfWidth,\n        predUpper: yh + predHalfWidth\n      });\n    }\n    return out;\n  }\n\n  return {\n    coef: [intercept, beta],\n    predict,\n    evaluateGrid,\n    confidenceInterval\n  };\n}\n\nexport default regressionLogistic;\n"]}

package/es/common/regression-lowess.d.ts

@@ -1,12 +1,18 @@
+/**
+ * 局部线性 LOWESS。默认全量观测、span=2/3、三次稳健重加权。
+ * delta 显式插值加速；maxSamples 仅保留显式选择的历史抽样用途。
+ * predict 与 evaluateGrid 查询构建时固定的同一模型，不触发再拟合。
+ */
 export declare function regressionLowess(data: any[], x?: (d: any) => number, y?: (d: any) => number, options?: {
     span?: number;
     degree?: 1 | 0;
     iterations?: number;
     alpha?: number;
     maxSamples?: number;
+    delta?: number;
 }): {
     predict: (x0: number | number[]) => number | number[];
-    evaluate: any;
+    evaluate: (x0: number | number[]) => number | number[];
     evaluateGrid: (N: number) => {
         x: number;
         y: number;
@@ -19,5 +25,9 @@ export declare function regressionLowess(data: any[], x?: (d: any) => number, y?
         predLower: number;
         predUpper: number;
     }[];
+    fitted: {
+        x: number;
+        y: number;
+    }[];
 };
 export default regressionLowess;