Finding Relation

Based on the graph, it appears that the peak in sales right before the end of the year is not random; the top 5 stores have the same pattern. Now, let's explore another parameter that we haven't examined yet. We will investigate whether there is a correlation between the temperature and shop income. We assume that colder temperatures outside will lead to fewer visitors and lower sales. To answer this question, we will create a scatter plot of temperatures vs. weekly sales. Since there are more than 6,000 records in the dataset, we will only use data for the top 5 best-selling stores.


              1234567891011121314151617
            
# Loading the libraries
import pandas as pd
import matplotlib.pyplot as plt
import seaborn as sns

# Reading the data
df = pd.read_csv('https://codefinity-content-media.s3.eu-west-1.amazonaws.com/72be5dde-f3e6-4c40-8881-e1d97ae31287/shops_data3.csv')
df['Date'] = pd.to_datetime(df['Date'], dayfirst = True)

# Preparing data
top_stores = [20, 4, 14, 13, 2]
data = df.loc[df['Store'].isin(top_stores)]

# Initializing a scatter plot
sns.scatterplot(x = 'Temperature', y = 'Weekly_Sales', data = data, alpha = 0.5)
# Displaying the plot
plt.show()

However, it seems that there is no clear correlation between temperature and weekly sales since most of the data points are located in the lower part of the plot. There are only a few data points with significantly higher sales in the upper part of the plot. What are these points? Let's find out.

¿Todo estuvo claro?

¡Gracias por tus comentarios!

Sección 4. Capítulo 3