qesuite 1.0.14 → 1.0.15

package/dist/index.d.mts

@@ -329,6 +329,10 @@ declare const Distributions: {
         */
         pdf(x: number, a: number, b: number): number;
     };
+    ChiSq: {
+        cdf(x: number, df: number): number;
+        RightTail(x: number, df: number): number;
+    };
     Exponential: {
         /**
         * Cumulative Distribution Function. Calculates the probability of a random variable  with a value less than or equal to x occuring from the exponential distribution with the specified scale parameter and optional threshold parameter.
@@ -588,6 +592,23 @@ declare const Distributions: {
         * @param scale [OPTIONAL] (σ): The scale parameter of the specified distribution. The scale parameter of the normal distribution is equivalent to the Standard Deviation.
         */
         pdf(x: number, location?: number, scale?: number): number;
+        RyanJoiner: {
+            Test(data: number[]): {
+                RJ: number;
+                critical: number;
+            };
+            Critical(data: number[]): number;
+        };
+        JarqueBera(data: number[]): {
+            JB: number;
+        };
+        Chebyshev(data: number[], spec: Specification): number;
+        Skewness(data: any): number;
+        Kurtosis(data: any): number;
+        KolmogorovSmirnov(data: any): {
+            KS: number;
+            p: any;
+        };
     };
     T: {
         /**

package/dist/index.d.ts

@@ -329,6 +329,10 @@ declare const Distributions: {
         */
         pdf(x: number, a: number, b: number): number;
     };
+    ChiSq: {
+        cdf(x: number, df: number): number;
+        RightTail(x: number, df: number): number;
+    };
     Exponential: {
         /**
         * Cumulative Distribution Function. Calculates the probability of a random variable  with a value less than or equal to x occuring from the exponential distribution with the specified scale parameter and optional threshold parameter.
@@ -588,6 +592,23 @@ declare const Distributions: {
         * @param scale [OPTIONAL] (σ): The scale parameter of the specified distribution. The scale parameter of the normal distribution is equivalent to the Standard Deviation.
         */
         pdf(x: number, location?: number, scale?: number): number;
+        RyanJoiner: {
+            Test(data: number[]): {
+                RJ: number;
+                critical: number;
+            };
+            Critical(data: number[]): number;
+        };
+        JarqueBera(data: number[]): {
+            JB: number;
+        };
+        Chebyshev(data: number[], spec: Specification): number;
+        Skewness(data: any): number;
+        Kurtosis(data: any): number;
+        KolmogorovSmirnov(data: any): {
+            KS: number;
+            p: any;
+        };
     };
     T: {
         /**

package/dist/index.js

@@ -975,6 +975,17 @@ var Distributions = {
       return Math.pow(x, a - 1) * Math.pow(1 - x, b - 1) / Beta.fn(a, b);
     }
   },
+  ChiSq: {
+    cdf(x, df) {
+      if (df === 2) {
+        return 1 - Math.exp(-x / 2);
+      }
+      return NaN;
+    },
+    RightTail(x, df) {
+      return 1 - Distributions.ChiSq.cdf(x, df);
+    }
+  },
   Exponential: {
     /**
     * Cumulative Distribution Function. Calculates the probability of a random variable  with a value less than or equal to x occuring from the exponential distribution with the specified scale parameter and optional threshold parameter.
@@ -1469,6 +1480,100 @@ var Distributions = {
     */
     pdf(x, location = 0, scale = 1) {
       return 1 / (scale * Math.sqrt(2 * Math.PI)) * Math.exp(-0.5 * Math.pow((x - location) / scale, 2));
+    },
+    RyanJoiner: {
+      Test(data) {
+        let N = data.length;
+        let Z = [];
+        let Y = data.sort(function(a, b) {
+          return a - b;
+        });
+        let numSum = 0;
+        let denSum = 0;
+        let mean = Mean(data);
+        let variance = Math.pow(StDev.S(data), 2);
+        for (let i = 0; i < N; i++) {
+          Z[i] = Distributions.Normal.inv((i + 1 - 0.375) / (N + 0.25));
+          numSum += Z[i] * (Y[i] - mean);
+          denSum += Math.pow(Z[i], 2);
+        }
+        let RJ = numSum / Math.sqrt(variance * (N - 1) * denSum);
+        return {
+          RJ,
+          critical: Distributions.Normal.RyanJoiner.Critical(data)
+        };
+      },
+      Critical(data) {
+        let N = data.length;
+        return 1.0063 - 0.6118 / N + 1.3505 / Math.pow(N, 2) - 0.1288 / Math.pow(N, 0.5);
+      }
+    },
+    JarqueBera(data) {
+      let N = data.length;
+      let S = Distributions.Normal.Skewness(data);
+      let K = Distributions.Normal.Kurtosis(data);
+      let JB = N / 6 * (Math.pow(S, 2) + 0.25 * Math.pow(K, 2));
+      return {
+        JB
+        // p: Distributions.ChiSq.RightTail(JB, 2)
+      };
+    },
+    Chebyshev(data, spec) {
+      let mean = Mean(data);
+      let standard_deviation = StDev.S(data);
+      let AbsUSL = spec.USL - mean;
+      let AbsLSL = mean - spec.LSL;
+      let min;
+      if (AbsUSL < AbsLSL) {
+        min = AbsUSL;
+      } else {
+        min = AbsLSL;
+      }
+      return min / (10 * standard_deviation);
+    },
+    Skewness(data) {
+      let N = data.length;
+      let mean = Mean(data);
+      let numsum = 0;
+      let densum = 0;
+      for (let i = 0; i < N; i++) {
+        let inner = data[i] - mean;
+        numsum += Math.pow(inner, 3);
+      }
+      let num = numsum * N;
+      let den = (N - 2) * (N - 1) * Math.pow(StDev.S(data), 3);
+      return num / den;
+    },
+    Kurtosis(data) {
+      let N = data.length;
+      let mean = Mean(data);
+      let numsum = 0;
+      let densum = 0;
+      for (let i = 0; i < N; i++) {
+        let inner = data[i] - mean;
+        numsum += Math.pow(inner, 4);
+      }
+      let num = numsum * N * (N + 1) / ((N - 1) * (N - 2) * (N - 3) * Math.pow(StDev.S(data), 4)) - 3 * Math.pow(N - 1, 2) / ((N - 2) * (N - 3));
+      return num;
+    },
+    KolmogorovSmirnov(data) {
+      let N = data.length;
+      data.sort();
+      let mean = Mean(data);
+      let std = StDev.S(data);
+      let dMinusArray = [];
+      let dPlusArray = [];
+      for (let i = 1; i <= N; i++) {
+        let Z_i = this.cdf(data[i - 1], mean, std);
+        dPlusArray.push(i / N - Z_i);
+        dMinusArray.push(Z_i - (i - 1) / N);
+      }
+      let KS = Math.max(Math.max(...dMinusArray), Math.max(...dPlusArray));
+      let p = GoodnessOfFit.AndersonDarling.p(KS, N);
+      return {
+        KS,
+        p
+      };
     }
   },
   T: {

package/dist/index.mjs

@@ -888,6 +888,17 @@ var Distributions = {
       return Math.pow(x, a - 1) * Math.pow(1 - x, b - 1) / Beta.fn(a, b);
     }
   },
+  ChiSq: {
+    cdf(x, df) {
+      if (df === 2) {
+        return 1 - Math.exp(-x / 2);
+      }
+      return NaN;
+    },
+    RightTail(x, df) {
+      return 1 - Distributions.ChiSq.cdf(x, df);
+    }
+  },
   Exponential: {
     /**
     * Cumulative Distribution Function. Calculates the probability of a random variable  with a value less than or equal to x occuring from the exponential distribution with the specified scale parameter and optional threshold parameter.
@@ -1382,6 +1393,100 @@ var Distributions = {
     */
     pdf(x, location = 0, scale = 1) {
       return 1 / (scale * Math.sqrt(2 * Math.PI)) * Math.exp(-0.5 * Math.pow((x - location) / scale, 2));
+    },
+    RyanJoiner: {
+      Test(data) {
+        let N = data.length;
+        let Z = [];
+        let Y = data.sort(function(a, b) {
+          return a - b;
+        });
+        let numSum = 0;
+        let denSum = 0;
+        let mean = Mean(data);
+        let variance = Math.pow(StDev.S(data), 2);
+        for (let i = 0; i < N; i++) {
+          Z[i] = Distributions.Normal.inv((i + 1 - 0.375) / (N + 0.25));
+          numSum += Z[i] * (Y[i] - mean);
+          denSum += Math.pow(Z[i], 2);
+        }
+        let RJ = numSum / Math.sqrt(variance * (N - 1) * denSum);
+        return {
+          RJ,
+          critical: Distributions.Normal.RyanJoiner.Critical(data)
+        };
+      },
+      Critical(data) {
+        let N = data.length;
+        return 1.0063 - 0.6118 / N + 1.3505 / Math.pow(N, 2) - 0.1288 / Math.pow(N, 0.5);
+      }
+    },
+    JarqueBera(data) {
+      let N = data.length;
+      let S = Distributions.Normal.Skewness(data);
+      let K = Distributions.Normal.Kurtosis(data);
+      let JB = N / 6 * (Math.pow(S, 2) + 0.25 * Math.pow(K, 2));
+      return {
+        JB
+        // p: Distributions.ChiSq.RightTail(JB, 2)
+      };
+    },
+    Chebyshev(data, spec) {
+      let mean = Mean(data);
+      let standard_deviation = StDev.S(data);
+      let AbsUSL = spec.USL - mean;
+      let AbsLSL = mean - spec.LSL;
+      let min;
+      if (AbsUSL < AbsLSL) {
+        min = AbsUSL;
+      } else {
+        min = AbsLSL;
+      }
+      return min / (10 * standard_deviation);
+    },
+    Skewness(data) {
+      let N = data.length;
+      let mean = Mean(data);
+      let numsum = 0;
+      let densum = 0;
+      for (let i = 0; i < N; i++) {
+        let inner = data[i] - mean;
+        numsum += Math.pow(inner, 3);
+      }
+      let num = numsum * N;
+      let den = (N - 2) * (N - 1) * Math.pow(StDev.S(data), 3);
+      return num / den;
+    },
+    Kurtosis(data) {
+      let N = data.length;
+      let mean = Mean(data);
+      let numsum = 0;
+      let densum = 0;
+      for (let i = 0; i < N; i++) {
+        let inner = data[i] - mean;
+        numsum += Math.pow(inner, 4);
+      }
+      let num = numsum * N * (N + 1) / ((N - 1) * (N - 2) * (N - 3) * Math.pow(StDev.S(data), 4)) - 3 * Math.pow(N - 1, 2) / ((N - 2) * (N - 3));
+      return num;
+    },
+    KolmogorovSmirnov(data) {
+      let N = data.length;
+      data.sort();
+      let mean = Mean(data);
+      let std = StDev.S(data);
+      let dMinusArray = [];
+      let dPlusArray = [];
+      for (let i = 1; i <= N; i++) {
+        let Z_i = this.cdf(data[i - 1], mean, std);
+        dPlusArray.push(i / N - Z_i);
+        dMinusArray.push(Z_i - (i - 1) / N);
+      }
+      let KS = Math.max(Math.max(...dMinusArray), Math.max(...dPlusArray));
+      let p = GoodnessOfFit.AndersonDarling.p(KS, N);
+      return {
+        KS,
+        p
+      };
     }
   },
   T: {

package/index.ts

@@ -1054,7 +1054,18 @@ export const Distributions = {
     pdf(x: number, a: number, b: number){
       return (Math.pow(x, a - 1) * Math.pow(1 - x, b - 1))/Beta.fn(a, b);
     }    
-  }, 
+  },
+  ChiSq: {
+    cdf(x: number, df: number){
+      if(df === 2){
+        return 1 - Math.exp(-x/2);
+      }
+      return NaN
+    },
+    RightTail(x: number, df: number){
+      return 1 - Distributions.ChiSq.cdf(x, df);
+    }
+  },
   Exponential: {
     /**
     * Cumulative Distribution Function. Calculates the probability of a random variable  with a value less than or equal to x occuring from the exponential distribution with the specified scale parameter and optional threshold parameter.
@@ -1546,8 +1557,107 @@ export const Distributions = {
     */
     pdf(x: number, location: number = 0, scale: number = 1) {
       return (1 / (scale * Math.sqrt(2 * Math.PI))) * Math.exp(-.5 * Math.pow((x - location) / scale, 2))
+    },
+    RyanJoiner: {
+      Test(data: number[]){
+        let N = data.length;
+    
+        let Z: number[] = [];
+        let Y = data.sort(function (a, b){return a-b});
+        let numSum = 0;
+        let denSum = 0;
+        let mean = Mean(data);
+        let variance = Math.pow(StDev.S(data), 2);
+    
+        for(let i = 0; i < N; i++){
+          Z[i] = Distributions.Normal.inv((i + 1 - 0.375)/ (N + 0.25))
+          numSum += Z[i]*(Y[i]-mean);
+          denSum += Math.pow(Z[i], 2);
+        }
+
+        let RJ = numSum/Math.sqrt(variance*(N-1)*denSum);
+        return {
+          RJ: RJ,
+          critical: Distributions.Normal.RyanJoiner.Critical(data)
+        }
+      },
+      Critical(data: number[]){
+        let N = data.length;
+        return 1.0063-(0.6118/N)+(1.3505/Math.pow(N, 2))-(.1288/Math.pow(N, .5))      
+      }
+    },
+    JarqueBera(data: number[]){
+      let N = data.length;
+      let S = Distributions.Normal.Skewness(data);
+      let K = Distributions.Normal.Kurtosis(data);
+      
+      let JB = (N/6)*(Math.pow(S, 2) + .25 * Math.pow(K, 2));
+      return {
+        JB: JB,
+        // p: Distributions.ChiSq.RightTail(JB, 2)
+      }
+    },
+    Chebyshev(data: number[], spec: Specification){  
+      let mean = Mean(data);
+      let standard_deviation =StDev.S(data);
+      let AbsUSL = spec.USL - mean;
+      let AbsLSL = mean - spec.LSL;
+      let min;
+      if(AbsUSL < AbsLSL){
+        min = AbsUSL;
+      }else{
+        min = AbsLSL;
+      }  
+      return min/(10*standard_deviation);
+    },
+    Skewness(data){
+      // SRC: https://www.omnicalculator.com/statistics/skewness#what-are-skewness-and-kurtosis
+      let N = data.length;
+      let mean = Mean(data);
+      let numsum = 0;
+      let densum = 0;
+      for(let i = 0; i < N; i++){
+        let inner = data[i] - mean;
+        numsum += Math.pow(inner, 3);
+      }
+      let num = numsum * N;
+      let den = (N - 2) * (N - 1) * Math.pow(StDev.S(data),3)
+      return num/den;
+    },
+    Kurtosis(data){
+      // SRC: https://www.omnicalculator.com/statistics/skewness#what-are-skewness-and-kurtosis
+      let N = data.length;
+      let mean = Mean(data);
+      let numsum = 0;
+      let densum = 0;
+      for (let i = 0; i < N; i++) {
+        let inner = data[i] - mean;
+        numsum += Math.pow(inner, 4);
+      }
+      let num = numsum * N * (N + 1) / ((N - 1) * (N - 2) * (N - 3) * Math.pow(StDev.S(data), 4)) - 3 * Math.pow(N - 1, 2) / ((N-2) * (N-3));
+      return num;
+    },
+    KolmogorovSmirnov(data){
+      let N = data.length;
+      data.sort();
+      let mean = Mean(data);
+      let std = StDev.S(data);
+      let dMinusArray: number[] = [];
+      let dPlusArray: number[] = [];
+      for(let i = 1; i <= N; i++){
+        let Z_i = this.cdf(data[i-1], mean, std)
+        dPlusArray.push((i/N)-Z_i);
+        dMinusArray.push(Z_i-(i-1)/N);
+
+      }
+      let KS = Math.max(Math.max(...dMinusArray),Math.max(...dPlusArray));
+      let p = GoodnessOfFit.AndersonDarling.p(KS, N);      
+      return {
+        KS: KS,
+        p: p
+      };
     }
-  },
+},
   T: {
     /**
     * Cumulative Distribution Function. Calculates the probability of a random variable  with a value less than or equal to x occuring from the T distribution with the specified degrees of freedom. (LEFT TAILED T-TEST)

package/package.json

@@ -1,6 +1,6 @@
 {
   "name": "qesuite",
-  "version": "1.0.14",
+  "version": "1.0.15",
   "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_15.md

@@ -0,0 +1,15 @@
+Added the following functionality to Distributions:
+
+- ChiSq
+- - cdf
+- - RightTail
+
+- Normal
+- - RyanJoiner
+- - - Test
+- - - Critical
+- - JarqueBera
+- - Chebyshev
+- - Skewness
+- - Kurtosis
+- - KolmogorovSmirnov