Recipe Objective

Pearson"s correlation is very important statical data that we need many times. We can calculate it manually but it takes time.

So this is the recipe on how we can determine Pearson"s correlation in Python

Step 1 - Importing Library

We have imported stats, seaborn and pandas which is needed.

Step 2 - Creating a dataframe

We have created a empty dataframe and then added rows to it with random numbers. df = pd.DataFrame() df["x"] = random.sample(range(1, 100), 75) df["y"] = random.sample(range(1, 100), 75) print(); print(df.head())

Step 3 - Calculating Pearsons correlation coefficient

We hawe defined a function with differnt steps that we will see.

• We have calculated mean and standard deviation of x and length of x
def pearson(x,y): n = len(x) standard_score_x = []; standard_score_y = []; mean_x = stats.mean(x) standard_deviation_x = stats.stdev(x)
• We atre calculating mean and standard deviation of y
mean_y = stats.mean(y) standard_deviation_y = stats.stdev(y)
• We are calculating standard score by dividing difference of observation and mean with standard deviation. We have done this for both X and Y
for observation in x: standard_score_x.append((observation - mean_x)/standard_deviation_x) for observation in y: standard_score_y.append((observation - mean_y)/standard_deviation_y) return (sum([i*j for i,j in zip(standard_score_x, standard_score_y)]))/(n-1)

Printing the Results

result = pearson(df.x, df.y) print() print("Pearson"s correlation coefficient is: ", result) sns.lmplot("x", "y", data=df, fit_reg=True) plt.show()

    x   y
0  96  62
1   1  81
2  27  73
3  55  26
4  83  93

Pearson"s correlation coefficient is:  -0.006387074440361877