Shulou(Shulou.com)06/01 Report--

This article introduces the relevant knowledge of "how to use python to analyze sales decline". In the operation process of actual cases, many people will encounter such difficulties. Next, let Xiaobian lead you to learn how to deal with these situations! I hope you can read carefully and learn something!

case review

Sales of rice balls declined

One of the pub's best-selling dishes, rice balls, has sold about 20 percent less in recent months than in the same period last year

The tavern gives 47 dishes, and sales figures for each day for three months (Question: What caused the decline in rice ball sales?)

Analysis of changes in dish sales

Store the data in csv format and import it into python. In order to visualize the time series of rice ball sales, draw the time series diagram of rice ball sales within 3 months.

menu. Date = pd.to_datetime(menu. Date)

import matplotlib.pyplot as plt

import pylab

menus.index = menus.iloc[:,1]

menu.loc[menu. name =='rice ball'].iloc[:,2].plot()

plt.ylabel("Sales")

pylab.show()

A time series plot, i.e. a plot whose horizontal axis represents the Timeline. As can be seen from the chart, after April, the overall trend of the broken line is declining, that is, sales decline. The line fluctuates sharply because sales go up over the weekend. Look at the sales of fried rice.

menu.loc[menu. name =='fried rice'].iloc[:,2].plot()

plt.ylabel("Sales")

pylab.show()

It can be seen intuitively from the chart that the sales of fried rice fluctuate equally badly, but there is no upward or downward trend as a whole. Look at the sales of noodles.

menu.loc[menu. name =='spaghetti'].iloc[:,2].plot(label='spaghetti')

menu.loc[menu. name =='fried noodles in sauce'].iloc[:,2].plot(label=' fried noodles in sauce')

menu.loc[menu. name =='udon'].iloc[:,2].plot(label=' udon')

menu.loc[menu. Name =='Mixed Noodles'].iloc[:,2].plot(label=' Mixed Noodles')

menu.loc[menu. name =='ramen'].iloc[:,2].plot(label='ramen')

plt.ylabel("Sales")

plt.legend()

pylab.show()

As you can see from the chart, sales of each of the five noodle varieties have risen a bit since April. Generally speaking, few people will start with rice noodles, so it is normal for one side to increase and the other side to decrease. Although it seems that the sales of rice balls are related to the sales of noodles, is it really because of noodles that the sales of rice balls are falling? There is a false correlation between the two, because they are not causal, but should be caused by some other reason to change the sales of these two. Through scatter plot, we can get the results of pairwise comparison of various foods and judge whether there is correlation. From the scatter plot, we can see that there is a significant negative correlation between rice balls and milk.

plt.scatter(menu.loc[menu. name =='rice ball'].iloc[:,2], menu.loc[menu. name ==' milk'].iloc[:,2])

plt.xlabel("rice ball")

plt.ylabel("milk")

pylab.show()

In real life, there is no substitution relationship between rice balls and milk, that is, there should not be a negative correlation between the two, so the situation shown in the image is different from the actual situation. Let's look at the milk sales chart.

menu.loc[menu. name =='milk'].iloc[:,2].plot()

plt.ylabel("Sales")

pylab.show()

As can be seen from the chart, milk sales have risen sharply since April. The investigation found that the tavern imported a kind of rice milk, namely porridge, which had a substitution relationship with rice balls, resulting in a decrease in rice ball sales.

A few small concepts.

Correlation coefficient: Looking at the scatter plot, if the points are banded from the bottom left to the top right, then the two data are positively correlated. If it is distributed from top left to bottom right, it is negatively correlated. But this is only a judgment based on the naked eye. Sometimes it is better to judge by numbers than by pictures. For example, calculate the correlation coefficient between milk and rice ball sales.

a = menu.loc[menu. name =='milk'].iloc[:,2]b = menu.loc[menu. name ==' rice ball'].iloc[:,2]c = pd.concat([a,b],axis=1)c.columns = ['milk',' rice ball'] c.corr()

The results are:

Milk Rice Ball Milk 1.000000-0.574642 Rice Ball-0.5746421.00000

The correlation coefficient represents the strength of the linear relationship between the two sets of data, and the value range is [-1, 1]. The closer to 1 means the higher the positive correlation, and the closer to-1 means the higher the negative correlation. From the calculation results, we can see that milk and rice balls are weakly negatively correlated.

"How to use python to analyze sales decline" content is introduced here, thank you for reading. If you want to know more about industry-related knowledge, you can pay attention to the website. Xiaobian will output more high-quality practical articles for everyone!

Welcome to subscribe "Shulou Technology Information " to get latest news, interesting things and hot topics in the IT industry, and controls the hottest and latest Internet news, technology news and IT industry trends.