Posts

Correlation analysis - Some caveats

Image
Correlation analysis is an integral part of many EDAs given that it is so easy to execute, visualize and interpret by technical and non-technical audeinces alike. It is a very intuitive metric to compute, and it also happens to be exceedingly useful. An introductory correlation analysis can help us understand what features may be interesting and deserve a closer look, point to potential multicollinearity concerns, and help in identifying opportunities for dimension reduction. BUT... there is always a but! As with any other metric, correlation comes with its caveats and gotchas which are helpful to keep in mind to get the most out of it - without misusing it. Here are a few pointers to note: 1. The most commonly used correlation method, Pearson Correlation, is a measure of the strength of the linear relationship between two variables. The keyword 'linear' is important; just because two variables do not have a high correlation coefficient does not mean that there is no relati

Predicting Customer Conversion: A Data-Driven Approach

Image
The following blog post reflects the efforts of my team - Daria Herasymova, Hirak Bhayani, and I - during Duke University's Datathon competition. As Business Analytics graduate students of Duke University’s Fuqua School of Business, we have a plethora of resources to explore & have access to many opportunities to build our data science skills and experience. We were excited to put to good use all that we learned in our classroom sessions conducted by Prof. Alex Belloni at an annual datathon that took place on campus. Organized by Duke Machine Learning, the Duke Datathon was a weekend-long event where teams solved an existent business challenge faced by a marketing technology & consumer engagement company that works with over 60,000 brands across a wide array of industries. The firm predicts customer churn by examining content across 80 billion web pages each day and deriving meaning from words and sentences to determine likely purchase intent through page