Kernel Density Estimation — A Gentle Introduction to Non-Parametric Statistics
Kernel Density Estimation is One of the Foundational Concepts in Non-Parametric Statistics. Let Me Take You on an Intuitive Ride Around This Topic.
Image credit: Me
Normality is a Myth!
Back in the 20th century, when Statistics was still in its infancy and computers weren’t that popular, it was norm to assume normality as the distribution from which data was generated. It was mostly because it made the calculations less tedious in the age when all results were hand calculated.
But with the rise of computational power these assumptions can safely be put aside and more insights can be drawn directly for the data. Even the availability of data in this Big Data era made Statisticians to adopt more modern techniques — Non-Parametric Statistics. Here we will discuss one such method to estimate the probability distribution, Kernel Density Estimation.
Read More at: Kernel Density Estimation — A Gentle Introduction to Non-Parametric Statistics
Rishi Dey Chowdhury
Master of Statistics
My research interests include Artificial Intelligence, Quant Trading, Deep Learning and their applications in Market Microstructure, Computer Vision and NLP.