def Pop_correlation_coefficient(my_population, my_population2):
mean1, mean2 = Mean(my_population), Mean(my_population2)
sd1, sd2 = Sd(my_population), Sd(my_population2)
correlation = 0
for i in range(0, len(my_population)):
factor_a = (my_population[i] - mean1) / sd1
factor_b = (my_population2[i] - mean2) / sd2
correlation = correlation + (factor_a * factor_b)
pop_correlation_coefficient = correlation / len(my_population)
return pop_correlation_coefficient
def Standardised_score(my_population):
my_mean = Mean(my_population)
my_sd = Sd(my_population)
standardised_score = list()
for x in my_population:
my_score = (x - my_mean) / my_sd
standardised_score.append(my_score)
return standardised_score
def Z_score(my_population):
my_mean = Mean(my_population)
my_sd = Sd(my_population)
zscore = list()
for x in my_population:
my_score = (x - my_mean) / my_sd
zscore.append(my_score)
return zscore
def Variance(my_population):
sum = 0
index = 0
my_variance = 0
calculated_mean = Mean(my_population)
for index in range(0, len(my_population)):
sum += ((my_population[index] - calculated_mean) ** 2)
if (len(my_population) == 0): my_variance = 0
else: my_variance = sum/len(my_population)
return my_variance
def Proportion(my_population):
# Any number 20% above or 20% below the mean is being considered as outlier.
# This function returns the proportion of the population which is outlier.
my_mean = Mean(my_population)
lower_limit = 0.8 * my_mean
higher_limit = 1.2 * my_mean
count = 0
proportion_success = list()
for i in my_population:
if (i>higher_limit) or (i< lower_limit):
count = count + 1
proportion_success.append(i)
p = (count/float(len(my_population)))
return p,proportion_success
def sample_mean(self,my_population):
sample_list = sample(my_population,math.floor(len(my_population)/2))
write_answer("answer_sample_mean.csv",Mean(sample_list))
return Sample_mean(sample_list)
def mean(self,my_population):
return round(Mean(my_population),2)