Evaluate the Chi square of observed dataset respect to a expected dataset.

[Statistics] - Chi square.c

// Coded by ScratchyCode
// Compile in gcc with option -lm
#include <stdio.h>
#include <stdlib.h>
#include <string.h>
#include <math.h>

// maximum size for file name
#define LEN 100

double chiSquare(long long int dim, double expected[], double observed[]);
double distChiSquare(long long int k, double array[]);
double average(long long int N, double array[]);
double sigma(long long int N, double average, double array[]);
int fileLines(char file[]);

int main(){
    long long int i, dim;
    char expected[LEN], observed[LEN];
    FILE *input;

    printf("\nEnter the file name (or path) with the expected data: ");
    fgets(expected,sizeof(expected),stdin);
    expected[strlen(expected)-1] = 0;
    
    printf("\nEnter the file name (or path) with the observed data: ");
    fgets(observed,sizeof(expected),stdin);
    observed[strlen(observed)-1] = 0;
    
    // file lines expected == file line observed
    dim = fileLines(expected);
    if(dim <= 5){
        printf("\nUnreliable test. Insufficient variables number.\n");
    }
    
    // arrays for expected data
    double *vect1x = calloc(dim,sizeof(double));
    double *vect1y = calloc(dim,sizeof(double));
    // arrays for observed data
    double *vect2x = calloc(dim,sizeof(double));
    double *vect2y = calloc(dim,sizeof(double));
    if(vect1x == NULL || vect1y == NULL || vect2x == NULL || vect2y == NULL){
        perror("\nError");
        exit(1);
    }
    
    // take the expected data
    input = fopen(expected,"r");
    if(input == NULL){
        perror("\nError");
        exit(1);
    }
    
    for(i=0; i<dim; i++){
        fscanf(input,"%lf %lf",&vect1x[i],&vect1y[i]);
    }
    
    fclose(input);
    
    // take the observed data
    input = fopen(observed,"r");
    if(input == NULL){
        perror("\nError");
        exit(1);
    }
    
    for(i=0; i<dim; i++){
        fscanf(input,"%lf %lf",&vect2x[i],&vect2y[i]);
    }
    
    fclose(input);
    
    printf("Computing...\n");
    
    // observed data average
    double x = average(dim,vect2y);
    // computing chi square
    double chi = chiSquare(dim,vect1y,vect2y);
    // calculate the density of probability
    double freeDegree = dim;
    double probab = (1 / (pow(2,(int)freeDegree/2) * tgamma(freeDegree))) * pow(x,(int)(freeDegree/2)-1) * exp(-(x/2));
    
    //printf("Have a probability of %lf%% to have a chi square of %lf\n",probab*100,chi);
    printf("Chi square: %lf\n",chi);
    
    free(vect1x);
    free(vect1y);
    free(vect2x);
    free(vect2y);
    
    return 0;
}

double chiSquare(long long int dim, double expected[], double observed[]){
    long long int i, freeDegree=dim-1;
    double sum=0;
    
    for(i=0; i<dim; i++){
        sum += (pow((expected[i] - observed[i]),2))/expected[i];
    }
    
    return sum;
}

double distChiSquare(long long int k, double array[]){
    long long int i;
    double sumSquare=0;
    
    for(i=0; i<k; i++){
        sumSquare += pow(array[i],2);
    }
    
    return sumSquare;
}

double average(long long int N, double array[]){
    long long int i;
    double sum = 0;
    
    for(i=0; i<N; i++){
        sum += array[i];
    }
    
    sum = (double)sum/N;
    
    return sum;
}

double sigma(long long int N, double average, double array[]){
    long long int i;
    double sigma, sum = 0;
    
    for(i=0; i<N; i++){
        sum += pow((array[i] - average),2);
    }
    
    sigma = sqrt(sum/(N-1));
    
    return sigma;
}

int fileLines(char file[]){
    int lines=0;
    char c;
    
    FILE *input = fopen(file,"r");
    if(input == NULL){
        perror("\nError");
        exit(1);
    }
    
    while((c = getc(input)) != EOF){
        if(c == '\n'){
            lines++;
        }
    }
    
    fclose(input);
    
    return lines;
}