Multivariate Analysis

In this segment, we will discuss the next part of EDA, i.e., multivariate analysis.

So far, you have learnt how two variables can be visualised based on their type, for example, numerical, categorical, etc. Now, let’s analyse two variables simultaneously. One of the key features of multivariate analysis is that it gives you a very precise idea about the various elements, since you are now combining multiple variables to visualise the data set. You will learn about this in more detail in the forthcoming videos.

First, let’s listen to Rahim as he explains his inferences from the bank marketing dataset in the next video.

In the video above, you saw that our expert performed a three-variable analysis between education, marital status and response. You can see that people who are married and who have completed just their primary education are least likely to give a positive response on term deposits. This can be explained by the fact that people educated only up to the primary level are not aware of the benefits of term investments.  Also, married individuals need money to fulfil their daily needs, and they require cash-on-hand to buy the daily essentials; hence, they won’t prefer investing in term deposits.

In the next video, you will see how job and marital status are varying with respect to the response variable.

In the video, you saw that the combinations of married with blue-collar, entrepreneur and housemaid are least likely to go for term deposits. The highest rate of positive response came from students with single marital status. The bank should, therefore, consider these aspects before taking any decision.

Having gone through all these examples, you must have a clear idea about the EDA process and the various steps involved in it.

Report an error