This model is implemented to predict the sales based on the budgeting spent on various platforms like marketing/advertisement
Businesses use regression models to understand how changes in a set of independent variables affect a dependent one.
For ecommerce businesses, the dependent variable is often sales. It can also be conversion rates or email signups, as examples. For this article, I’ll assume the dependent variable is sales. The independent variables could be email sends and expenditures on social media and search engine optimization, as examples. The regression model lets business owners measure, one at a time, each independent variable’s impact on sales.
In other words, a regression model can predict, say, how much a 20 percent increase in Facebook ad spend will increase sales. It can use past sales and, perhaps, weather data by date to predict how a coming storm will slow or speed sales. It can also give you an idea of the increase or decrease in sales resulting from additional email sends — a decrease would indicate subscriber annoyance.
A simple regression formula could be:
Y = A+B(X)
- Y is the dependent variable — sales, email signups.
- X is the value of the independent variable — Facebook ads, email frequency.
- B is a constant that reflects how much Y changes for every value of X. (Getting an accurate number may require a mathematician or an app.)
- A is a constant that equals the value of Y when X is zero. Determine A by plugging 0 into X
In this project, A Linear Regression model has been implemented to predict sales from marketing/advertising spends on different platforms such as TV, Radio and Newspaper.