Topics in Quantitative Finance: Machine Learning for Finance (2017-18 Module 3)

Announcements

• [NEW 4.02] Group formation, Project proposal (4.12 W7 Thur), Presentation (5.7 Mon 7PM) (Project List)
• [NEW 3.19] Mid-term exam will be on 4.9 (W7 Monday, Rm 401)
• [3.7] Class mailing list is created as PHBS.TQF@allmail.net. If you did not receive a test e-mail, let me know.
• [2.26] Email is the preferred method of communication. Mail list will be set up soon.

Course Note

Lectures:

• 18 (04.26 Thur): ML-related thesis presentation by two students
• 17
• 16
• 15 (04.16 Mon): PML Ch. 11 Unsupervised learning, Ch. 12 Neural Networks
• 14 (04.12 Thurs): Project proposal (Project List),
• 13 (04.09 Mon): Mid-term exam (Rm 401) Solution
• 12 (04.04 Wed): PML Ch. 7 Bagging/RF/AdaBoosting, Ch 8 Quick Glimpse
• 11 (04.02 Mon): PML Ch. 6 Confusion matrix, F1 score, ROC AUC, Ch 7
• 10 (03.29 Thur): PML Ch. 6 Bias-variance tradeoff, Cross validation
• 09 (03.26 Mon): PML Ch. 5 Feature extraction (SVD/PCA and LDA)
• 08 (03.22 Thur): PML Ch. 4 (Data preprocessing, Feature selection)
• 07 (03.19 Mon): PML Ch. 3 (Kernel SVM, KNN, Decision Tree), Slides (SVM, Tree)
• 06 (03.15 Thur): PML Ch. 3 (Logistic Regression, Regularization, SVM), Slides (SVM)
• 05 (03.12 Mon): PML Ch. 2 (Perceptron, Adaline, Gradient descent), Slides (Weight update)
• 04 (03.08 Thur): Slides (Vector Matrix Notations, Linear/Logistic Regression), ISLR-python Ch. 3
• 03 (03.05 Mon): Python Crash Course (continued), Slides (Intro)
• 02 (03.01 Thur): Python Notebook, Github Desktop, Python Crash Course
• 01 (02.26 Mon): Course overview (Syllabus), Python, Github, Etc.

Homeworks:

• Set 1 [Due by 3.5]:

  • Register on Github.com and let TA know your ID. Give your full name in your profile.
  • Accept invitation to 2017.TQF-ML team from TA
  • Install Python Aanconda and Github Desktop. Clone PHBS/python-machine-learning-book-2nd-edition and run code/ch01/ch01.ipynb Send screenshots to TA

• Set 2 [Due by 3.29]

  • Create your own GitHub repository for homework GITHUB_ID/PHBS_TQFML (make sure to make it public)
  • Bank Marketing Data Set: UCI, Download
  • Write a Jupyter notebook GITHUB_ID/PHBS_TQFML/HW/bank_marketing.ipynb:
    • load data (bank.csv, smaller sample), normalize, and devide training/test sets
    • randomly select 2 or 3 features
    • apply the methods covered in Ch. 3 with SK-learn (logistic regress, SVM, decision tree, etc)
    • check the accuracy and plot the outcome
    • repeat above to find better feature
    • commit the best result and don't foget to sync to the repository

• Project proposal [Due by the end of 4.11]

  • Email repository name, team members, data set and brief plan to Professor and TA
  • Designate one repository GITHUB_ID/PHBS_TQFML/Project and create README.md file to have the information above

Classes:

• Lectures: Monday & Thursday 8:30 AM – 10:20 AM
• Venue: PHBS Building, Room 229

Instructor: Jaehyuk Choi

• Office: PHBS Building, Room 755
• Phone: 86-755-2603-0568
• Email: jaehyuk@phbs.pku.edu.cn
• Office Hour: Monday & Thursday 1:30 – 2:30 PM or by appointment

Teaching Assistance: Chenru LIU (刘晨茹)

• Email: 1501213463@sz.pku.edu.cn
• TA Office Hour: Tuesday & Friday 1-2 PM (Room 213/214)

Course overview

The purpose of Topics in Quantitative Finance is to introduce students to recent trends and advanced research topics in quantitative methods of business and finance. This year’s course is dedicated to machine learning (ML) for finance. ML has been one of the hottest technology in software engineering. This course will explore the possibility of applying ML to finance and business. The course will give students the basic ideas and intuition behind the popular ML methods and hands-on experience of using ML software package such as SK-learn and Tensorflow (Google). Each student is required to complete a final course project.

Prerequisites

There is no formal prerequisites. However, undergraduate-level knowledge in probability/statistics and previous experience in programming language is highly recommended.

Textbooks and Reading Materials

• PML (primary textbook): Python Machine Learning

learning-book-2nd-edition) by Sebastian Raschka

• ISLR: An Introduction to Statistical Learning (with Applications in R) by James, Witten, Hastie, and Tibshirani
• CML: Coursera Machine Learning taught by Andrew Ng
• DL: Deep Learning by Goodfellow, Bengio, and Courville
• JPM-AI: Big Data and AI Strategies by J.P. Morgan

Useful Github Repositories

• PML: PHBS/python-machine-learning-book-2nd-edition (forked)
• ISLR-Python: PHBS/ISLR-python (forked) ISRL implemented in Python

Assessment / Grading Details

• Attendance 20%, Mid-term Exam 20% (New), Assignments 20%, Final Project 40%
• Mid-term exam: 4.9 (W7 Mon)
• Project proposal: 4.12 (W7 Thur)
• Presentation (5.7 Mon 7PM)
• Exams are open-book without computer/phone/calculator use
• You may form a group of up to 2 people for course project. Extra credit will be given to individual projects.
• Grade in letters (e.g., A+, A-, ... ,D+, D, F). A- or above < 30% and B- or above < 90%.