Altair RISE

A program to recognize and reward our most engaged community members

Nominate Yourself Now!

If I spend extra money in marketing cost will i get more revenue?

If I spend extra money in marketing cost will i get more revenue?

How should I use Linear regression to determine whether the marketing cost will affect the revenue with using rapid miner?

Or is there other ways to determine whether the marketing cost will affect the revenue with using rapid miner?

Data from excel is attach below, anyone can help me to solve?

Find more posts tagged with

AI Studio

Linear Regression

Accepted answers

BalazsBaranyRM

Hi,

this is quite straightforward. You just import or open your data (using Read Excel), set the target attribute to label (using Set Role) and connect this to a Linear Regression operator. The model description will then tell you if marketing cost has a positive or negative influence on the revenue.

This is the theory. In practice you have a very small dataset, just a few attributes, and in the crazy economy of the last few years you could easily find an artefact in your data that isn't there in reality. So don't base important decisions on this model as long as the caveats I mentioned are valid.

Regards,
Balázs

All comments

BalazsBaranyRM

Wonderer

@BalazsBarany thanks for your reply. Just a short recap on the guides you provide for me.

Process in Rapid Miner
Excel file insert in Rapid Miner--- Select attribute (Sales and Marketing)--- Set Role (Sales and Marketing) as label --- Linear Regression --- end.
Did i get it right base on your advices?

Or can I this way with the RMP provide below?

(Rapidminer marketing towards revenue).rmp

BalazsBaranyRM

Hi @Wonderer,

you would usually model the revenue explained by the other variables. You selected the marketing expenses as the label. This would build a model for explaining that from the revenue which is not how I interpret your question.

Please watch a few introductory videos about modeling in the Academy. You will then better understand the ideas behind machine learning.

Regards,
Balázs

Wonderer

@BalazsBarany so the label I should put is Revenue not marketing as label, right?

BalazsBaranyRM

Exactly. The label is the attribute you will predict in the future.

In a business context, your goal would be to model the future outcome of something that is dependent on things you can control. You can choose to invest more or less into marketing hoping to influence the revenue. It doesn't work the other way around.

Regards,
Balázs