How to determine the coefficient of influence of the weather?

Question

Algorithms

Hello!



In the screenshot, revenue by days with actual weather, the better the weather, the lower the revenue and vice versa

. How can you determine how the weather affects revenue and, based on this, take into account the weather forecast when planning revenue for a specific day?

Answer 1

First of all. The "influence" of one parameter on another can be estimated using the correlation coefficient. In this case - most likely - Prison's correlation coefficient.
Secondly, the "influence" of one parameter on another can be revealed in the form of a regression model.
The correlation coefficient simply indicates whether it is possible to ASSUME that the parameters are related to each other. The regression approach tries to build models. But in order for this model to be adequate, that is, in order to be able to really predict something with its help, more serious restrictions must be imposed on the data. Whether they are performed on your data - you need to check.
Another thing is that in reality everything is much more complicated. The data you have (assume) sales volume depending on the actual temperature on that day. And you are going to build models according to the forecast. And it's not the same at all. And - if you take it seriously - it is much more difficult than a direct task.
And besides, do not forget that your data most likely also has a seasonal component, and the word "seasonal" here can mean 7 days, and 30 days and 365 days. And if everything is done "in an adult way", then time series models are already starting to work.