lifeEvents = read.csv("life_events.csv");
# store data set
counter <- numeric(25);
# counter for valid responses for each life event
totalHappiness <- numeric(25);
# Added up all of the happiness levels for all of the respondents who answered.
average <- numeric(25);
for(j in 28:52){
# Cycles through all of the life events.
  for (i in 1:10054) {
# Cycles through all of the respondents.
    if (lifeEvents[i,j] > -1)
    {
      counter [j-27] <- counter[j-27] + 1;
# Increments the counter, j-27 to get back to the correct row of column, because life event data starts in the 28th row.
      totalHappiness[j-27] <- totalHappiness[j-27] + lifeEvents[i,j];
# Increments total happiness. j-27 for the same reason.      
      
    }
    
  }
}


for(k in 1:25){
  
  average[k]= totalHappiness[k]/counter[k];
#Standard average function. Total divided by the number of people.
}

lifeEvents = read.csv("life_events.csv");
# store data set
counter <- numeric(25);
# counter for valid responses for each life event
peoplePerLevel <- matrix(0, nrow = 10, ncol = 25);
# Counts number of people per happiness level - 10 rows by 25 columns
distributionTable <- matrix(0, nrow = 10, ncol = 25);
# Distribution for each happiness level - 10 rows by 25 columns
hellingerSum <- matrix(0, nrow = 25, ncol = 25);
# Data frame to store the Hellinger distances for each life event.

#This double for loop calculates the number of people for each happiness level, for each life event.
for(j in 28:52){
#
  for (i in 1:10054){
# Cycles through all of the respondents
    if(lifeEvents[i,j] > -1){
# Eliminates all of the missing responses in the counter variable      
      counter[j-27] <- counter[j-27] + 1;
# Assigns the counter number found to the appropriate slot in the array. (Which life event)      
      if(lifeEvents[i,j] == 1) {
        peoplePerLevel[1,j-27] <- peoplePerLevel[1,j-27] + 1;
        
      }
# If a respondent gave 1 for life event j, in the matrix peoplePerLevel is incremented by 1. J will cycle from 1-25.   
      if(lifeEvents[i,j] == 2) {
        peoplePerLevel[2,j-27] <- peoplePerLevel[2,j-27] + 1;
        
      }    
# If a respondent gave 2 for life event j, in the matrix peoplePerLevel is incremented by 1. J will cycle from 1-25.
      if(lifeEvents[i,j] == 3) {
        peoplePerLevel[3,j-27] <- peoplePerLevel[3,j-27] + 1;
        
      }     
# If a respondent gave 3 for life event j, in the matrix peoplePerLevel is incremented by 1. J will cycle from 1-25.
      if(lifeEvents[i,j] == 4) {
        peoplePerLevel[4,j-27] <- peoplePerLevel[4,j-27] + 1;
# If a respondent gave 4 for life event j, in the matrix peoplePerLevel is incremented by 1. J will cycle from 1-25.        
      }   
      if(lifeEvents[i,j] == 5) {
        peoplePerLevel[5,j-27] <- peoplePerLevel[5,j-27] + 1;
# If a respondent gave 5 for life event j, in the matrix peoplePerLevel is incremented by 1. J will cycle from 1-25.
      }
      if(lifeEvents[i,j] == 6) {
        peoplePerLevel[6,j-27] <- peoplePerLevel[6,j-27] + 1;
# If a respondent gave 6 for life event j, in the matrix peoplePerLevel is incremented by 1. J will cycle from 1-25.
      }
      if(lifeEvents[i,j] == 7) {
        peoplePerLevel[7,j-27] <- peoplePerLevel[7,j-27] + 1;
# If a respondent gave 7 for life event j, in the matrix peoplePerLevel is incremented by 1. J will cycle from 1-25.       
      }
      if(lifeEvents[i,j] == 8) {
        peoplePerLevel[8,j-27] <- peoplePerLevel[8,j-27] + 1;
# If a respondent gave 8 for life event j, in the matrix peoplePerLevel is incremented by 1. J will cycle from 1-25.      
      }
      if(lifeEvents[i,j] == 9) {
        peoplePerLevel[9,j-27] <- peoplePerLevel[9,j-27] + 1;
# If a respondent gave 9 for life event j, in the matrix peoplePerLevel is incremented by 1. J will cycle from 1-25.        
      }
      if(lifeEvents[i,j] == 10) {
        peoplePerLevel[10,j-27] <- peoplePerLevel[10,j-27] + 1;
        
      }
# If a respondent gave 1 for life event j, in the matrix peoplePerLevel is incremented by 1. J will cycle from 1-25.
      
      
      
      
    }
  }
}

#This double for loop calculates the percentage of people for each happiness level, for each life event. 
for(j in 1:25){
  
  for(i in 1:10){
    
    distributionTable[i,j] <- peoplePerLevel[i,j]/counter[j];
    
    
    
  }
  
  
}
# Takes the number of people who had a happiness level of i for life event j, and divides it by the total number of people for life event j.
for(j in 1:25){
  
  k=j+1;
# Makes the starting number for the second life event of the pair of life events.
  while (k <= 25){
  
    for(i in 1:10){
    
    hellingerSum[j,k] <- hellingerSum[j,k] + (sqrt(distributionTable[i,j]) - sqrt(distributionTable[i,k]))^2
# Equation of the Hellinger Distance - Take the sqrt of percentage of people who chose level "i" for life event j
# subtract it from percentage of people who chose level "i" for life event k.
  }
  k=k+1;
# Increment k by 1 to generate a new pair. ie, 1,2 to 1,3
  }
  
}

hellingerSum <- 1/sqrt(2)*sqrt(hellingerSum)
# Take the final sqrt and multiply by 1/sqrt 2. Do this to the whole equation.