Jeremiah Lowin Machine Learning in Investing [Invest Like the Best, EP.105]
b'My guest this week is one of my best and oldest friends, Jeremiah Lowin. Jeremiah has had a fascinating career, starting with advanced work in statistics before moving into the risk management field in the hedge fund world. Through his career he has studied data, risk, statistics, and machine learning\\u2014the last of which is the topic of our conversation today.\\xa0\\n He has now left the world of finance to found a company called Prefect, which is a framework for building data infrastructure. Prefect was inspired by observing frictions between data scientists and data engineers, and solves these problems with a functional API for defining and executing data workflows. These problems, while wonky, are ones I can relate to working in quantitative investing\\u2014and others that suffer from them out there will be nodding their heads. In full and fair disclosure, both me and my family are investors in Jeremiah\\u2019s business.\\n You won\\u2019t have to worry about that potential conflict of interest in today\\u2019s conversation, though, because our focus is on the deployment of machine learning technologies in the realm of investing. What I love about talking to Jeremiah is that he is an optimist and a skeptic. He loves working with new statistical learning technologies, but often thinks they are overhyped or entirely unsuited to the tasks they are being used for. We get into some deep detail on how tests are set up, the importance of data, and how the minimization of error is a guiding light in machine learning and perhaps all of human learning, too. Let\\u2019s dive in.\\n For more episodes go to InvestorFieldGuide.com/podcast.\\n Sign up for the book club, where you\\u2019ll get a full investor curriculum and then 3-4 suggestions every month at InvestorFieldGuide.com/bookclub.\\n Follow Patrick on Twitter at @patrick_oshag\\n Show Notes\\n 2:06 - (First Question) \\u2013 What do people need to think about when considering using machine learning tools\\n 3:19 \\u2013 Types of problems that AI is perfect for\\n 6:09 \\u2013 Walking through an actual test and understanding the terminology\\n 11:52 \\u2013 Data in training: training set, test set, validation set\\n 13:55 \\u2013 The difference between machine learning and classical academic finance modelling\\n 16:09 \\u2013 What will the future of investing look like using these technologies\\n 19:53 \\u2013 The concept of stationarity\\n 21:31 \\u2013 Why you shouldn\\u2019t take for granted label formation in tests\\n 24:12 \\u2013 Ability for a model to shrug\\n 26:13 \\u2013 Hyper parameter tuning\\n 28:16 \\u2013 Categories of types of models\\n 30:49 \\u2013 Idea of a nearest neighbor or K-Means Algorithm\\n 34:48 \\u2013 Trees as the ultimate utility player in this landscape\\n 38:00 \\u2013 Features and data sets as the driver of edge in Machine Learning\\n 40:12 \\u2013 Key considerations when working through time series\\n 42:05 \\u2013 Pitfalls he has seen when folks try to build predictive market investing models\\n 44:36 \\u2013 Getting started\\n 46:29 \\u2013 Looking back at his career, what are some of the frontier vs settled applications of machine learning he has implemented\\n 49:49 \\u2013 Does intereptability matter in all of this\\n 52:31 \\u2013 How gradient decent fits into this whole picture \\xa0\\n \\xa0\\n Learn More\\n For more episodes go to InvestorFieldGuide.com/podcast.\\xa0\\n Sign up for the book club, where you\\u2019ll get a full investor curriculum and then 3-4 suggestions every month at InvestorFieldGuide.com/bookclub\\n Follow Patrick on twitter at @patrick_oshag'