qesuite 1.0.60 → 1.0.61

package/dist/index.d.mts

@@ -1248,8 +1248,8 @@ declare const Correlation: {
 };
 declare function Count(arr: any[], value: any): number;
 declare function DescriptiveStatistics(data: number[]): {
-    Mean: number;
     N: number;
+    Mean: number;
     StDev: number;
     SE: number;
 };
@@ -1349,8 +1349,8 @@ declare const EquivalenceTesting: {
             Summary: HTMLCanvasElement;
         };
         DescriptiveStatistics: {
-            Mean: number;
             N: number;
+            Mean: number;
             StDev: number;
             SE: number;
         };
@@ -1387,8 +1387,8 @@ declare const EquivalenceTesting: {
             Summary: HTMLCanvasElement;
         };
         DescriptiveStatistics: {
-            Mean: number;
             N: number;
+            Mean: number;
             StDev: number;
             SE: number;
             Variable: any;
@@ -1427,8 +1427,8 @@ declare const EquivalenceTesting: {
             Summary: HTMLCanvasElement;
         };
         DescriptiveStatistics: {
-            Mean: number;
             N: number;
+            Mean: number;
             StDev: number;
             SE: number;
             Variable: any;
@@ -1517,8 +1517,9 @@ declare const NonparametricTesting: {
             p: number;
         };
     };
-    KruskallWallis(data: number[][]): {
+    KruskalWallis(data: any[][]): {
         DescriptiveStatistics: {
+            Sample: any;
             N: number;
             Median: any;
             MeanRank: any;

package/dist/index.d.ts

@@ -1248,8 +1248,8 @@ declare const Correlation: {
 };
 declare function Count(arr: any[], value: any): number;
 declare function DescriptiveStatistics(data: number[]): {
-    Mean: number;
     N: number;
+    Mean: number;
     StDev: number;
     SE: number;
 };
@@ -1349,8 +1349,8 @@ declare const EquivalenceTesting: {
             Summary: HTMLCanvasElement;
         };
         DescriptiveStatistics: {
-            Mean: number;
             N: number;
+            Mean: number;
             StDev: number;
             SE: number;
         };
@@ -1387,8 +1387,8 @@ declare const EquivalenceTesting: {
             Summary: HTMLCanvasElement;
         };
         DescriptiveStatistics: {
-            Mean: number;
             N: number;
+            Mean: number;
             StDev: number;
             SE: number;
             Variable: any;
@@ -1427,8 +1427,8 @@ declare const EquivalenceTesting: {
             Summary: HTMLCanvasElement;
         };
         DescriptiveStatistics: {
-            Mean: number;
             N: number;
+            Mean: number;
             StDev: number;
             SE: number;
             Variable: any;
@@ -1517,8 +1517,9 @@ declare const NonparametricTesting: {
             p: number;
         };
     };
-    KruskallWallis(data: number[][]): {
+    KruskalWallis(data: any[][]): {
         DescriptiveStatistics: {
+            Sample: any;
             N: number;
             Median: any;
             MeanRank: any;

package/dist/index.js

@@ -4354,7 +4354,7 @@ var Distributions = {
       } else {
         min = AbsLSL;
       }
-      return min / (10 * standard_deviation);
+      return min / standard_deviation;
     },
     JarqueBera(data) {
       let N = data.length;
@@ -4894,9 +4894,10 @@ var Distributions = {
       };
       let t = void 0;
       try {
-        t = table[`${p}`][df - 1];
         if (p < 0.5) {
-          t = -t;
+          t = -1 * table[`${1 - p}`][df - 1];
+        } else {
+          t = table[`${p}`][df - 1];
         }
         return t;
       } catch (err) {
@@ -6672,7 +6673,6 @@ var MANOVA = {
           return factors[inter][i];
         }).join("*"));
       }
-      console.log(factor);
       let q = Unique(factor).length - 1;
       let E = MANOVA.ErrorMatrix(responses, factor);
       let H = MANOVA.HypothesisMatrix(responses, factor);
@@ -6742,10 +6742,8 @@ var MANOVA = {
         let factorSSCP = Matrix.Multiply(Matrix.Transpose(rDevMatrix), rDevMatrix);
         ErrorMatrix = Matrix.Sum(ErrorMatrix, factorSSCP);
       });
-      console.log({ DeviationMatrix, nMatrix });
       let devXn = Matrix.Scale(DeviationMatrix, nMatrix);
       let HypothesisMatrix = Matrix.Multiply(Matrix.Transpose(DeviationMatrix), devXn);
-      console.log({ HypothesisMatrix, ErrorMatrix });
       let EStarStar_1_xH = Matrix.Multiply(Matrix.Inverse(ErrorMatrix), HypothesisMatrix);
       let EigenValues = Matrix.EigenValues(EStarStar_1_xH);
       let eigenWilk = Product(EigenValues.map((l) => {
@@ -7152,8 +7150,8 @@ function Combinatrix(values) {
 }
 function DescriptiveStatistics(data) {
   return {
-    Mean: Mean(data),
     N: data.length,
+    Mean: Mean(data),
     StDev: StDev.S(data),
     SE: StandardError(data)
   };
@@ -7831,10 +7829,11 @@ var NonparametricTesting = {
       }
     };
   },
-  KruskallWallis(data) {
+  KruskalWallis(data) {
     let combinedData = [];
     data.forEach((g) => {
-      combinedData.push(...g);
+      let { name: groupName, data: groupData } = FormatInputData(g);
+      combinedData.push(...groupData);
     });
     let ranks = Rank.Combined(data);
     let avgRanks = ranks.map((r) => {
@@ -7863,6 +7862,7 @@ var NonparametricTesting = {
     return {
       DescriptiveStatistics: data.map((group, i) => {
         return {
+          Sample: FormatInputData(group).name ?? `Group ${i + 1}`,
           N: group.length,
           Median: Median(group),
           MeanRank: avgRanks[i],
@@ -11457,7 +11457,6 @@ function ANOMEChartEMP(operators, parts, responses, title, confidenceLevel = 0.0
   let g = Unique(parts).length * byTreatments.length;
   let m = rData.length / g;
   let d2 = UnbiasingConstant.D2prime(g, m);
-  console.log(d2);
   let A\u03B1 = t\u03B1_2 / d2 * Math.sqrt(1 / n);
   let LDL = XbarBar - A\u03B1 * Rbar;
   let UDL = XbarBar + A\u03B1 * Rbar;
@@ -11503,7 +11502,6 @@ function ANOMRChartEMP(operators, parts, responses, title, confidenceLevel = 0.0
   byTreatments.forEach((t, ti) => {
     let { name: tName, data: tData } = FormatInputData(t);
     let rng = Mean(tData);
-    console.log({ rng, tData });
     dataset.values.push(new Point(ti, rng));
     chartSettings.axis.x.customGridlines.push(new CustomGridline(ti, tName));
   });
@@ -11810,7 +11808,6 @@ function GageLinearityAndBias(parts, referenceValues, responses, processVariatio
       let \u03C3 = (Math.max(...biases) - Math.min(...biases)) / d2;
       t = bias / (\u03C3 / Math.sqrt(n2));
     }
-    console.log(t, df);
     let pValue = Distributions.T.cdf(-Math.abs(t), df) + Distributions.T.RightTail(Math.abs(t), df);
     partBiases.push({
       ReferenceValue: ref,

package/dist/index.mjs

@@ -4228,7 +4228,7 @@ var Distributions = {
       } else {
         min = AbsLSL;
       }
-      return min / (10 * standard_deviation);
+      return min / standard_deviation;
     },
     JarqueBera(data) {
       let N = data.length;
@@ -4768,9 +4768,10 @@ var Distributions = {
       };
       let t = void 0;
       try {
-        t = table[`${p}`][df - 1];
         if (p < 0.5) {
-          t = -t;
+          t = -1 * table[`${1 - p}`][df - 1];
+        } else {
+          t = table[`${p}`][df - 1];
         }
         return t;
       } catch (err) {
@@ -6546,7 +6547,6 @@ var MANOVA = {
           return factors[inter][i];
         }).join("*"));
       }
-      console.log(factor);
       let q = Unique(factor).length - 1;
       let E = MANOVA.ErrorMatrix(responses, factor);
       let H = MANOVA.HypothesisMatrix(responses, factor);
@@ -6616,10 +6616,8 @@ var MANOVA = {
         let factorSSCP = Matrix.Multiply(Matrix.Transpose(rDevMatrix), rDevMatrix);
         ErrorMatrix = Matrix.Sum(ErrorMatrix, factorSSCP);
       });
-      console.log({ DeviationMatrix, nMatrix });
       let devXn = Matrix.Scale(DeviationMatrix, nMatrix);
       let HypothesisMatrix = Matrix.Multiply(Matrix.Transpose(DeviationMatrix), devXn);
-      console.log({ HypothesisMatrix, ErrorMatrix });
       let EStarStar_1_xH = Matrix.Multiply(Matrix.Inverse(ErrorMatrix), HypothesisMatrix);
       let EigenValues = Matrix.EigenValues(EStarStar_1_xH);
       let eigenWilk = Product(EigenValues.map((l) => {
@@ -7026,8 +7024,8 @@ function Combinatrix(values) {
 }
 function DescriptiveStatistics(data) {
   return {
-    Mean: Mean(data),
     N: data.length,
+    Mean: Mean(data),
     StDev: StDev.S(data),
     SE: StandardError(data)
   };
@@ -7705,10 +7703,11 @@ var NonparametricTesting = {
       }
     };
   },
-  KruskallWallis(data) {
+  KruskalWallis(data) {
     let combinedData = [];
     data.forEach((g) => {
-      combinedData.push(...g);
+      let { name: groupName, data: groupData } = FormatInputData(g);
+      combinedData.push(...groupData);
     });
     let ranks = Rank.Combined(data);
     let avgRanks = ranks.map((r) => {
@@ -7737,6 +7736,7 @@ var NonparametricTesting = {
     return {
       DescriptiveStatistics: data.map((group, i) => {
         return {
+          Sample: FormatInputData(group).name ?? `Group ${i + 1}`,
           N: group.length,
           Median: Median(group),
           MeanRank: avgRanks[i],
@@ -11331,7 +11331,6 @@ function ANOMEChartEMP(operators, parts, responses, title, confidenceLevel = 0.0
   let g = Unique(parts).length * byTreatments.length;
   let m = rData.length / g;
   let d2 = UnbiasingConstant.D2prime(g, m);
-  console.log(d2);
   let A\u03B1 = t\u03B1_2 / d2 * Math.sqrt(1 / n);
   let LDL = XbarBar - A\u03B1 * Rbar;
   let UDL = XbarBar + A\u03B1 * Rbar;
@@ -11377,7 +11376,6 @@ function ANOMRChartEMP(operators, parts, responses, title, confidenceLevel = 0.0
   byTreatments.forEach((t, ti) => {
     let { name: tName, data: tData } = FormatInputData(t);
     let rng = Mean(tData);
-    console.log({ rng, tData });
     dataset.values.push(new Point(ti, rng));
     chartSettings.axis.x.customGridlines.push(new CustomGridline(ti, tName));
   });
@@ -11684,7 +11682,6 @@ function GageLinearityAndBias(parts, referenceValues, responses, processVariatio
       let \u03C3 = (Math.max(...biases) - Math.min(...biases)) / d2;
       t = bias / (\u03C3 / Math.sqrt(n2));
     }
-    console.log(t, df);
     let pValue = Distributions.T.cdf(-Math.abs(t), df) + Distributions.T.RightTail(Math.abs(t), df);
     partBiases.push({
       ReferenceValue: ref,

package/index.ts

@@ -4482,7 +4482,7 @@ export const Distributions = {
       }else{
         min = AbsLSL;
       }  
-      return min/(10*standard_deviation);
+      return min/(standard_deviation);
     },
     JarqueBera(data: number[]){
       let N = data.length;
@@ -5041,9 +5041,10 @@ export const Distributions = {
       }
       let t: any = undefined
       try {
-        t = table[`${p}`][df-1]
         if(p < 0.5){
-          t = -t
+          t = -1 * table[`${1 - p}`][df-1]
+        }else{          
+          t = table[`${p}`][df-1]
         }
         return t 
       }catch (err){
@@ -5770,7 +5771,7 @@ export const ANOVA = {
         let this_n = data[key].length;
         let standardError = summaryTable.ANOVA.Error.AdjMS/this_n;
         let tInv = Distributions.T.inv(0.05, summaryTable.ANOVA.Error.DF);
-
+        
         let CI_table = {
           name: key,
           N: this_n,
@@ -7015,7 +7016,6 @@ export const MANOVA = {
       for(let i = 0; i < responses[rKeys[0]].length; i++){
         factor.push(int.map(inter => {return factors[inter][i]}).join("*"))
       }
-      console.log(factor);
   
       let q = Unique(factor).length - 1;
       let E = MANOVA.ErrorMatrix(responses, factor);
@@ -7093,10 +7093,9 @@ export const MANOVA = {
         ErrorMatrix = Matrix.Sum(ErrorMatrix, factorSSCP); 
       })
 
-      console.log({DeviationMatrix, nMatrix})
       let devXn = Matrix.Scale(DeviationMatrix, nMatrix);
       let HypothesisMatrix = Matrix.Multiply(Matrix.Transpose(DeviationMatrix), devXn)//, responses[rKeys[0]].length/factorGroups.length)
-      console.log({HypothesisMatrix, ErrorMatrix})
+
       let EStarStar_1_xH = Matrix.Multiply(Matrix.Inverse(ErrorMatrix), HypothesisMatrix)
       let EigenValues: any = Matrix.EigenValues(EStarStar_1_xH);
       let eigenWilk = Product(EigenValues.map(l => {return 1/(1+l)}))
@@ -7529,8 +7528,8 @@ function Combinatrix(values: any[][]){
 
 export function DescriptiveStatistics(data: number[]){
   return {
-    Mean: Mean(data),
     N: data.length,
+    Mean: Mean(data),
     StDev: StDev.S(data),
     SE: StandardError(data)
   }
@@ -8232,10 +8231,11 @@ export const NonparametricTesting = {
       }
     }
   },
-  KruskallWallis(data: number[][]){
+  KruskalWallis(data: any[][]){
     let combinedData:any = [];
     data.forEach(g => {
-      combinedData.push(...g)
+      let {name: groupName, data: groupData} = FormatInputData(g)
+      combinedData.push(...groupData)
     })
     
     let ranks = Rank.Combined(data);
@@ -8257,6 +8257,7 @@ export const NonparametricTesting = {
     return {
       DescriptiveStatistics: data.map((group, i) => {
         return {
+          Sample: FormatInputData(group).name ?? `Group ${i+1}`,
           N: group.length,
           Median: Median(group),
           MeanRank: avgRanks[i],
@@ -12274,7 +12275,7 @@ export function ANOMEChartEMP(operators: any[], parts: any[], responses: any[],
   let g = Unique(parts).length * byTreatments.length;
   let m = rData.length / g;
   let d2 = UnbiasingConstant.D2prime(g, m) 
-  console.log(d2)
+  
   let Aα = (tα_2/d2) * Math.sqrt(1/n)
   let LDL = XbarBar - Aα * Rbar
   let UDL = XbarBar + Aα * Rbar
@@ -12326,7 +12327,7 @@ export function ANOMRChartEMP(operators: any[], parts: any[], responses: any[],
   byTreatments.forEach((t, ti) => {
     let {name: tName, data: tData} = FormatInputData(t);
     let rng = Mean(tData);
-    console.log({rng, tData})
+    
     dataset.values.push(new Point(ti, rng));
     chartSettings.axis.x.customGridlines.push(new CustomGridline(ti, tName))
   })
@@ -12662,7 +12663,7 @@ export function GageLinearityAndBias(parts: any[], referenceValues: number[], re
       let σ = (Math.max(...biases) - Math.min(...biases)) / d2;
       t = bias / (σ/Math.sqrt(n))
     }
-    console.log(t, df)
+    
     let pValue = Distributions.T.cdf(-Math.abs(t), df) + Distributions.T.RightTail(Math.abs(t), df)
 
     partBiases.push({

package/package.json

@@ -1,6 +1,6 @@
 {
   "name": "qesuite",
-  "version": "1.0.60",
+  "version": "1.0.61",
   "description": "Performs advanced statistical analysis of data. Specifically designed for engineering statistical analysis",
   "main": "dist/index.js",
   "module": "dist/index.mjs",

package/versions/1_0_61.md

@@ -0,0 +1,4 @@
+- Changed Order of Descriptive Statistics so that N is before Mean
+- Kruskal-Wallis now handles sample names
+- Fixed Bug in Distributions.T.Inv() to allow use of the invTable for p < 0.5
+- Changed Chebyshev to return k instead of k/10