• $50

Survey Course: Machine Learning in Investment Management

  • Course
  • 67 Lessons

In this course, we will introduce you to machine learning in investment management through four targeted case studies. Each case is designed to help you improve one step in the investment management pathway. This is all about giving you practical skills and showing you how experts do this in the real world!

You can watch the first lecture for free! (See the video above).

What You Will Learn

  • Master Hierarchical Clustering: Identify hidden relationships between asset classes using advanced clustering algorithms, giving you a new edge in asset class due diligence and investment strategy development.

  • Decode Model Behavior with Shapley Values: Gain deep insights into the driving forces behind your models and discover how to better interpret and trust your results by using Shapley values for clearer explanations.

  • Enhance Factor Models with Lasso Regression: Improve your understanding of investment behavior by applying Lasso regression to sharpen factor models, helping you make more informed, data-driven investment decisions.

  • Optimize Tactical Asset Allocation with Deep Learning: Learn how to leverage cutting-edge deep learning techniques to anticipate market shifts and make smarter allocation decisions between bonds and equities.

Contents

Course Introduction

In this section, we will introduce you to the course outline and what you will learn. We will also introduce you to our teaching method which includes quizzes, and teaching you how to think critically about what you will be learning.

READ THIS BEFORE YOU START THE COURSE
    00 - Introduction
    • 9 mins
    • 24.8 MB
    Important! A Brief Note on Retrieving Returns

      Case Study 1: Hierarchical Clustering for Asset Class Analysis

      In this case study, we show you how clustering analysis can help you understand how asset classes behave and if various asset classes behave in similar ways according to various risk and return metrics.

      Case Study 1: Slides
      • 209 KB
      Case Study 1: Python Notebook
      • 412 KB
      01 - Theory, Part 1
      • 17 mins
      • 49.3 MB
      02 - Theory, Part 2
      • 16 mins
      • 43.2 MB
      03 - Imports
      • 11 mins
      • 42.2 MB
      04 - Initialization
      • 11 mins
      • 44.3 MB
      05 - Feature Engineering, Part 1
      • 12 mins
      • 41.1 MB
      06 - Feature Engineering, Part 2
      • 23 mins
      • 71.8 MB
      07 - Generating the Feature Matrix
      • 7 mins
      • 24.1 MB
      08 - Hierarchical Clustering
      • 13 mins
      • 42.7 MB
      09 - Summary
      • 7 mins
      • 21.5 MB
      Case Study 1 Quiz
        Case Study 1: Deep Thinking Questions
        • 53.8 KB

        Case Study 2: Understanding Shapley Values

        In this case study, we will show you how powerful Shapley values can be for identifying feature importance in machine learning models. We will apply these values to our results from the first case study.

        Please make sure to download the needed files provided in this section.

        Case Study 2: Shapley Values - Slides
        • 278 KB
        Case Study 2: Python Notebook
        • 280 KB
        feature_matrix_scaled_final.csv
        • 7.06 KB
        mergings.pkl
        • 1.68 KB
        10 - Theory
        • 15 mins
        • 47 MB
        11 - Imports
        • 9 mins
        • 25.2 MB
        12 - Assigning Clusters
        • 8 mins
        • 27.4 MB
        13 - LightGBM
        • 7 mins
        • 31.5 MB
        14 - Shapley Values, Part 1
        • 19 mins
        • 70.2 MB
        15 - Shapley Values, Part 2
        • 13 mins
        • 48.4 MB
        16 - Visualization
        • 8 mins
        • 24.7 MB
        17 - Summary
        • 12 mins
        • 56.3 MB
        Case Study 2: Quiz
          Case Study 2: Detailed Answer for Question 10 of Multiple Choice.pdf
          • 65.5 KB
          Case Study 2: Deep Thinking Questions
          • 77.6 KB

          Case Study 3: Improving Factor Models with Lasso Regression

          In this case study, we demonstrate how you can create improved returns-based factor models by using an advanced regression technique called LASSO.

          Case Study 3: Lasso Regression - Python Notebook
          • 203 KB
          Case Study 3: Lasso Regression - Slides
          • 211 KB
          18 - Theory, Part 1
          • 13 mins
          • 35.5 MB
          19 - Theory, Part 2
          • 12 mins
          • 34.2 MB
          20 - Imports
          • 5 mins
          • 14.1 MB
          21 - Retreiving Fund Returns
          • 14 mins
          • 52.1 MB
          22 - Retreiving Factor Returns
          • 11 mins
          • 38 MB
          23 - Variance Inflation Factor and OLS
          • 18 mins
          • 66.7 MB
          24 - OLS Evaluation Metrics
          • 13 mins
          • 51.2 MB
          25 - LASSO Regression
          • 14 mins
          • 50.1 MB
          26 - LASSO Evaluation Metrics
          • 9 mins
          • 44.3 MB
          27 - AIC and BIC
          • 24 mins
          • 70.7 MB
          28 - Visualizing the Differences
          • 20 mins
          • 70.6 MB
          29 - Retreiving the Differences
          • 10 mins
          • 43.9 MB
          30 - Summary
          • 7 mins
          • 29.2 MB
          Case Study 3 Quiz
            Case Study 3: Deep Thinking Questions
            • 66.9 KB

            Case Study 4: Tactical Asset Allocation with Deep Learning

            In our fourth case study, we will demonstrate how you can assess the probability of SPY going up or down the next day using deep learning. This is a great example of how we can alter our asset allocation based on signals derived from machine learning.

            Case Study 4: Slides
            • 241 KB
            Case Study 4: Python Notebook
            • 104 KB
            31 - Theory, Part 1
            • 18 mins
            • 49.6 MB
            32 - Theory, Part 2
            • 13 mins
            • 35 MB
            33 - Imports
            • 10 mins
            • 25.8 MB
            34 - Feature Engineering
            • 19 mins
            • 67.8 MB
            35 - Creating the Target
            • 10 mins
            • 40.3 MB
            36 - Creating the Model, Part 1
            • 20 mins
            • 58.8 MB
            37 - Creating the Model, Part 2
            • 12 mins
            • 42.7 MB
            38 - Running the Model
            • 21 mins
            • 78.3 MB
            Case Study 4 Quiz
              Case Study 4: Deep Thinking Questions
              • 74 KB

              Course Summary

              This short video ties everything together and shows how the case studies affect the portfolio matrix and investment management workflow.

              39 - Course Summary
              • 13 mins
              • 38.1 MB
              Course Project

                Files

                This section contains the files needed for this course. Please make sure to watch the tutorial video on how to use the main_functions.py file and how to import them.

                DHI0008_functions.py
                • 18.3 KB
                DHI0008_functions_no_openbb.py
                • 15.8 KB
                requirements.txt
                • 185 Bytes

                Assignment - Hierarchical Clustering of Fixed Income Funds

                In this project, you will reinforce what you learned in case study 1 of our Introduction to Machine Learning in Portfolio Construction course.
                You will classify various fixed income funds based on different risk and return features.

                Fixed Income Fund Clustering - Project Notebook
                • 282 KB
                Fixed Income Fund Clustering - Project Solutions
                • 284 KB

                • $50

                Survey Course: Machine Learning in Investment Management

                • Course
                • 67 Lessons

                In this course, we will introduce you to machine learning in investment management through four targeted case studies. Each case is designed to help you improve one step in the investment management pathway. This is all about giving you practical skills and showing you how experts do this in the real world!