Correlations / predictive analytics for a car dealership

qlikview
correlation
predictive_model

#1

Greetings,
One of my clients (a car dealership) has asked me to help them with finding correlations based on their Car Sales data and Car Service data.
What would you think will be the best solution for that?
I believe it can be done by using the Pearson formula to find correlations and maybe integrating it in their ERP-system for future use. Or maybe there is a program for that that you guys could recommend (I had an idea about associative mechanisms of Qlik View/Qlik Sense to discover correlations, but I am not sure if it is a valid option because the client wants a SaaS-based instrument).
Please, share your piece of mind on the most efficient way of solving the correlations issue. Thanks.


#2

Hi @Alexy.

Please help me with how your data actually looks like so that I can assist you better.


#3

Hi @Alexy,

Azure Machine Learning (from Microsoft) is a cloud based platform where you can perform different types of data processing tasks, run machine learning algorithms and evaluate them. The platform is easy to use as it has a intuitive graphical user interface with drag & drop functionality. It is free of cost - all you need is a Microsoft id.

With specific reference to correlations, you can use the “Compute Linear Correlation” moduel in Azure ML. Connect your dataset to this module and it will give you the correlation values for all numerical variables. You can also integrate R & Python scripts into Azure ML. Check it out.