Deep Learning for Finance

Chapter 2. Essential Probabilistic Methods for Deep Learning

The rise and accessibility of technology have made it possible for everyone to deploy machine learning and deep learning algorithms for data analysis and optimization. But unfortunately, a large number of users do not understand the basics of the different learning models. This makes machine learning nothing short of a mystery box to them, which is a recipe for disaster.

Understanding fundamental concepts in probability, statistics, and math is essential for understanding and mastering data as well as for creating models that seek to interpret and forecast data. This chapter presents the basics of probability that are either directly or indirectly related to the algorithms. Note that you are unlikely to use these probability concepts in your everyday life, but it’s important to know where some algorithms draw their assumptions from.

A Primer on Probability

Probability is all about describing random variables and random events. The world is filled with randomness, and the best way to find your way through chaos is to try to explain it using probabilistic methods. Granted, the phrase explain chaos may be an oxymoron, as chaos cannot really be explained, but we humans cannot relinquish control over uncertain events. This is why we have developed tools to make sense out of our scary world.

You may wonder what is the use of understanding the basics of probability when trying to develop machine learning algorithms for financial ...