Implementazioni di algoritmi/Test Chi Quadrato

Wikibooks, manuali e libri di testo liberi.
Jump to navigation Jump to search
Indice del libro

Con Test Chi Quadrato si intende uno dei test di verifica d'ipotesi usati in statistica che utilizzano la variabile casuale Chi Quadrato per verificare se l'ipotesi nulla è probabilisticamente compatibile con i dati. A seconda delle ipotesi di partenza usate per costruire il test, tali test vengono considerati a volte parametrici e altre volte non parametrici.

Implementazione in Java[modifica]

      /**
         * Calculates the chi-square value for N positive integers less than r
         * Source: "Algorithms in C" - Robert Sedgewick - pp. 517
         * NB: Sedgewick recommends: "...to be sure, the test should be tried a few times,
         * since it could be wrong in about one out of ten times."
         * @param  randomNums  a uniformly-randomly-generated array of integers
         * @param  r           upper bound for the random range
         * @return             whether it is likely to be uniformly randomly generated
         */
        public static boolean isRandom(int[] randomNums, int r)
        {
                //According to Sedgewick: "This is valid if N is greater than about 10r"
                if (randomNums.length <= 10 * r)
                        return false;
 
                //PART A: Get frequency of randoms
                Map<Integer,Integer> ht = getFrequencies(randomNums);
 
                //PART B: Calculate chi-square - this approach is in Sedgewick
                double n_r = (double)randomNums.length / r;
                double chiSquare = 0;
 
                for (int v : ht.values())
                {
                        double f = v - n_r;
                        chiSquare += f * f;
                }
                chiSquare /= n_r;
 
                //PART C: According to Swdgewick: "The statistic should be within 2(r)^1/2 of r
                //This is valid if N is greater than about 10r"
                return Math.abs(chiSquare - r) <= 2 * Math.sqrt(r);
        }
 
 
        /**
         * @param  nums  an array of integers
         * @return       a Map, key being the number and value its frequency
         */
        private static Map<Integer,Integer> getFrequencies(int[] nums)
        {
                Map<Integer,Integer> freqs = new HashMap<Integer,Integer>();
 
                for (int x : nums)
                {
                        if (freqs.containsKey(x))
                                freqs.put(x, freqs.get(x) + 1);
                        else
                                freqs.put(x, 1);
                }
 
                return freqs;
        }

Implementazione in Python[modifica]

import math # for sqrt
 
def isRandom(randomNums, r):
        """ Calculates the chi-square value for N positive integers less than r
            Source: "Algorithms in C" - Robert Sedgewick - pp. 517
            NB: Sedgewick recommends: "...to be sure, the test should be tried a few times,
            since it could be wrong in about one out of ten times." """
        #Calculate the number of samples - N
        N = len(randomNums)
 
        #According to Sedgewick: "This is valid if N is greater than about 10r"
        if N <= 10 * r:
                return False
 
        N_r = N / r
        chi_square = 0
 
        #PART A: Get frequency of randoms
        HT = getFrequencies(randomNums)
 
        #PART B: Calculate chi-square - this approach is in Sedgewick
        for v in HT.itervalues():
                f = v - N_r
                chi_square += f * f
 
        chi_square /= N_r
 
        #PART C: According to Swdgewick: "The statistic should be within 2(r)^1/2 of r
        #This is valid if N is greater than about 10r"
        return abs(chi_square - r) <= 2 * math.sqrt(r)
 
 
def getFrequencies(nums):
        """ Gets the frequency of occurance of a randomly generated array of integers
            Output: A dictionary, key being the random number and value its frequency """
        freqs = {}
 
        for x in nums:
                if x in freqs:
                        freqs[x] += 1
                else:
                        freqs[x] = 1
 
        return freqs