sqlmath 0.0.1 → 2021.11.20

package/test.slr.mjs

@@ -1,212 +0,0 @@
-/*jslint beta, devel*/
-(function () {
-    function assertOrThrow(passed, msg) {
-
-// This function will throw <msg> if <passed> is falsy.
-
-        if (!passed) {
-            throw (
-                (msg && msg.stack)
-                ? msg
-                : new Error(msg)
-            );
-        }
-    }
-
-    function simpleLinearRegressionCreate(size = 8192) {
-
-// This function will create an object for calculating simple-linear-regression
-// statistics.
-// https://en.wikipedia.org/wiki/Simple_linear_regression
-
-        let nn = 0;
-        let sxx = 0;
-        let sxy = 0;
-        let syy = 0;
-        let xavg = 0;
-        let xx2 = 0;
-        let xyList = new Float64Array(size * 2);
-        let yavg = 0;
-
-        function pass1(xx, yy) {
-
-// This function will calculate simple-linear-regression in one pass using
-// Wellford's algorithm.
-// For a new value xx, compute the new nn, new xavg, the new sxx.
-// xavg accumulates the xavg of the entire dataset
-// sxx aggregates the squared distance from the xavg
-// nn aggregates the number of samples seen so far
-
-            let dx;
-            let dy;
-            xyList[2 * nn] = xx;
-            xyList[2 * nn + 1] = yy;
-            nn += 1;
-            xx2 += xx * xx;
-
-// calculate sxx
-
-            dx = xx - xavg;
-            xavg += dx / nn;
-            sxx += dx * (xx - xavg);
-
-// calculate syy
-
-            dy = yy - yavg;
-            yavg += dy / nn;
-            syy += dy * (yy - yavg);
-
-// calculate sxy
-
-            sxy += dx * (yy - yavg);
-        }
-
-        function pass2(xyList2) {
-
-// This function will calculate simple-linear-regression in two passes.
-
-            let ii;
-            let nn2;
-            let xx;
-            let yy;
-            nn = xyList2.length / 2;
-            nn2 = xyList2.length;
-
-// pass #1
-
-            ii = 0;
-            while (ii < nn2) {
-                xx = xyList2[ii];
-                yy = xyList2[ii + 1];
-                xyList[ii] = xx;
-                xyList[ii + 1] = yy;
-                xavg += xx;
-                yavg += yy;
-                xx2 += xx * xx;
-                ii += 2;
-            }
-            xavg = xavg / nn;
-            yavg = yavg / nn;
-
-// pass #2
-
-            ii = 0;
-            while (ii < nn2) {
-                xx = xyList[ii] - xavg;
-                yy = xyList[ii + 1] - yavg;
-                sxx += xx * xx;
-                syy += yy * yy;
-                sxy += xx * yy;
-                ii += 2;
-            }
-        }
-
-        function result() {
-            let alpha;
-            let beta;
-            let ii = 0;
-            let nn2 = nn * 2;
-            let salpha2;
-            let sbeta2;
-            let see = 0;
-            beta = sxy / sxx;
-            alpha = yavg - beta * xavg;
-            while (ii < nn2) {
-                see += (alpha + beta * xyList[ii] - xyList[ii + 1]) ** 2;
-                ii += 2;
-            }
-            see = see / (nn - 2);
-            sbeta2 = see / sxx;
-            salpha2 = sbeta2 * xx2 / nn;
-            return {
-                alpha,
-                beta,
-                correlation: sxy / Math.sqrt(sxx * syy),
-                nn,
-                salpha2,
-                sbeta2,
-                see,
-                sxx,
-                sxy,
-                syy,
-                xavg,
-                yavg
-            };
-        }
-
-        return {
-            pass1,
-            pass2,
-            result
-        };
-    }
-
-// test
-
-    (function () {
-        let slr1;
-        let xxList;
-        let yyList;
-
-        function round(num, decimal) {
-            return Math.round(num / decimal) * decimal;
-        }
-
-        function validate() {
-            let state = slr1.result();
-            console.error(state);
-            assertOrThrow(round(state.beta, 0.001) === 61.272, state.beta);
-            assertOrThrow(round(state.alpha, 0.001) === -39.062, state.alpha);
-            assertOrThrow(round(state.see, 0.0001) === 0.5762, state.see);
-            assertOrThrow(round(state.sbeta2, 0.0001) === 3.1539, state.sbeta2);
-            assertOrThrow(
-                round(state.salpha2, 0.00001) === 8.63185,
-                state.salpha2
-            );
-            assertOrThrow(
-                round(state.correlation, 0.0001) === 0.9946,
-                state.correlation
-            );
-        }
-
-        xxList = [
-            1.47, 1.50, 1.52, 1.55, 1.57, 1.60, 1.63, 1.65,
-            1.68, 1.70, 1.73, 1.75, 1.78, 1.80, 1.83
-        ];
-        yyList = [
-            52.21, 53.12, 54.48, 55.84, 57.20, 58.57, 59.93, 61.29,
-            63.11, 64.47, 66.28, 68.10, 69.92, 72.19, 74.46
-        ];
-
-// validate pass1
-
-        slr1 = simpleLinearRegressionCreate(xxList.length);
-        xxList.forEach(function (xx, ii) {
-            slr1.pass1(xx, yyList[ii]);
-        });
-        validate();
-
-// validate pass2
-
-        slr1 = simpleLinearRegressionCreate(xxList.length);
-        slr1.pass2(xxList.flatMap(function (xx, ii) {
-            return [
-                xx, yyList[ii]
-            ];
-        }));
-        validate();
-
-// validate pass1 and pass2
-
-        slr1 = simpleLinearRegressionCreate(xxList.length);
-        slr1.pass2(xxList.slice(2).flatMap(function (xx, ii) {
-            return [
-                xx, yyList[ii + 2]
-            ];
-        }));
-        xxList.slice(0, 2).forEach(function (xx, ii) {
-            slr1.pass1(xx, yyList[ii]);
-        });
-        validate();
-    }());
-}());