Mahalanobis Distance with the matrix libraries of MQ , and beer.

[ad_1]

First what’s it ?

Secondly some docs

Wikipedia web page

We are going to want :

Covariance matrix Matrix inversion Matrix multiplication Vector imply

The matrix and vectors library mql5 has covers all of our wants truly.

Okay so we are going to construct a construction that we will prep as soon as and reuse for so long as the pattern set will not be altering.

What will we imply with pattern set ?

A set of observations with properties .

To simplify , shall we say you’ve 10 candles (chart candles) ,and so they have OHLC . So you’ve 10 samples and 4 properties or 4 options.

Right here is an instance utilization

double open[],excessive[],low[],shut[];
ArrayResize(open,10,0);
ArrayResize(excessive,10,0);
ArrayResize(low,10,0);
ArrayResize(shut,10,0);
for(int i=0;i<10;i++){
   open[i]=iOpen(_Symbol,_Period,i+1);
   excessive[i]=iHigh(_Symbol,_Period,i+1);
   low[i]=iLow(_Symbol,_Period,i+1);
   shut[i]=iClose(_Symbol,_Period,i+1);
   }

mahalanober M;
M.setup(4,10);

  M.fill_feature(0,open);
  M.fill_feature(1,excessive);
  M.fill_feature(2,low);
  M.fill_feature(3,shut);


    double md=M.distanceOfSampleToDistribution(2);
    Print(“Mahalabonis Distance of bar 2 to the distribution “+DoubleToString(md,4));

    md=M.distanceOfSampleToSample(5,0);
    Print(“Mahalabonis Distance of bar[0] to bar[5] within the distribution “+DoubleToString(md,4));

and right here is the construction

struct mahalanober{

       non-public:
vector options[];
  bool stuffed[];
vector feature_means;
matrix covariance_matrix_inverse;
   int total_features,total_samples;
       public:
       mahalanober(void){reset();}
      ~mahalanober(void){reset();}
  void reset(){
       total_features=0;
       total_samples=0;
       ArrayFree(options);
       ArrayFree(stuffed);
       feature_means.Init(0);
       covariance_matrix_inverse.Init(0,0);
       }
  void setup(int _total_features,
             int _total_samples){
       total_features=_total_features;
       total_samples=_total_samples;
       ArrayResize(options,total_features,0);
       ArrayResize(stuffed,total_features,0);
       ArrayFill(stuffed,0,total_features,false);
       feature_means.Init(total_features);
       for(int i=0;i<ArraySize(options);i++){
          options[i].Init(total_samples);
          }
       }
  bool fill_feature(int which_feature_ix,
                    double &values_across_samples[]){
       if(which_feature_ix<ArraySize(options)){
       if(ArraySize(values_across_samples)==total_samples){
       for(int i=0;i<total_samples;i++){
          options[which_feature_ix][i]=values_across_samples[i];
          }
       feature_means[which_feature_ix]=options[which_feature_ix].Imply();
       stuffed[which_feature_ix]=true;

         if(all_filled()){
           calculate_inverse_covariance_matrix();
           }
       return(true);
       }else{
       Print(“MHLNB::fill_feature::Quantity of values doesn’t match whole samples”);
       }
       }else{
       Print(“MHLNB::fill_feature::Function(“+IntegerToString(which_feature_ix)+“) doesn’t exist”);
       }
       return(false);
       }
double distanceOfSampleToDistribution(int which_sample){
       if(all_filled()){
       if(which_sample<total_samples){

         matrix term0;
         term0.Init(total_features,1);
         for(int i=0;i<total_features;i++){
            term0[i][0]=options[i][which_sample]-feature_means[i];
            }

         matrix term3=term0;
         matrix term1;

           term1=term0.Transpose();

           matrix term2=term1.MatMul(covariance_matrix_inverse);

           matrix last_term=term2.MatMul(term3);


           return(MathSqrt(last_term[0][0]));
         }else{
         Print(“MLHNB::distanceOfSampleToDistribution()::Pattern (“+IntegerToString(which_sample)+“) doesn’t exist returning 0.0”);
         }
       }else{
       list_unfilled(“distanceOfSampleToDistribution()”);
       }
       return(0.0);
       }
double distanceOfSampleToSample(int sample_a,int sample_b){
       if(all_filled()){
       if(sample_a<total_samples){
       if(sample_b<total_samples){

         matrix term0;
         term0.Init(total_features,1);
         for(int i=0;i<total_features;i++){
            term0[i][0]=options[i][sample_a]-features[i][sample_b];
            }

         matrix term3=term0;
         matrix term1;

           term1=term0.Transpose();

           matrix term2=term1.MatMul(covariance_matrix_inverse);

           matrix last_term=term2.MatMul(term3);


           return(MathSqrt(last_term[0][0]));
           }else{
           Print(“MLHNB::distanceOfSampleToSample()::Pattern (“+IntegerToString(sample_b)+“) doesn’t exist returning 0.0”);
           }
         }else{
         Print(“MLHNB::distanceOfSampleToSample()::Pattern (“+IntegerToString(sample_a)+“) doesn’t exist returning 0.0”);
         }
       }else{
       list_unfilled(“distanceOfSampleToSample()”);
       }
       return(0.0);
       }
       non-public:
  void calculate_inverse_covariance_matrix(){

       matrix samples_by_features;
       samples_by_features.Init(total_samples,total_features);


         for(int f=0;f<total_features;f++){

           for(int s=0;s<total_samples;s++){
              samples_by_features[s][f]=options[f][s];
              }
           }

       matrix covariance_matrix=samples_by_features.Cov(false);

       covariance_matrix_inverse=covariance_matrix.Inv();
       }
  bool all_filled(){
       if(total_features>0){
       for(int i=0;i<total_features;i++){
          if(!stuffed[i]){
            return(false);
            }
          }
       return(true);
       }
       return(false);
       }
  void list_unfilled(string fx){
       for(int i=0;i<total_features;i++){
          if(!stuffed[i]){
            Print(“MLHNB::”+fx+“::Function(“+IntegerToString(i)+“) will not be stuffed!”);
            }
          }
       }
};