While being a vibrant subfield of computer science, machine learning is used for drawing models and methods from statistics, algorithms, computational complexity, control theory and artificial intelligence. It focuses on efficient algorithms for inferring good predictive models from large data sets and is natural candidate for problems arising in HFT – both trade execution & alpha generation.
In quantitative finance inference of models of predictive nature using historical data is obviously not new. Some examples include the coefficient estimation for CAPM, Fama and French factors. The granularity of data arising in HFT poses special challenges for machine learning. Often data microstructure at the resolution of individual orders, executions, hidden liquidity and cancellation including lack of understanding of how such granular data relates to actionable circumstances, namely profitably buying or selling shares, optimally executing a large order, etc.
Looking at the complexities mentioned above in machine learning, it is particularly important if one is interested in becoming a quantitative trader or researcher to learn machine learning.
Free Courses in Machine Learning
Perhaps the best introduction to machine learning is this highly-rated course by Stanford on Coursera. The course is taken by Professor Andrew Ng, who is praised for his ability to explain mathematical concepts involved in different areas of machine learning. The course gives a good introduction to machine learning, datamining and statistical pattern recognition. It requires the students to implement both Neural Networks and Vector machine (support vector machine to be precise). This course provides an actual hands-on training, and covers almost everything except new concepts like deep learning. This course by Prof Ng is definitely our pick for beginners!
Want questions like – Can machines learn? How exactly do they learn? An introductory course by CalTech teaches machine learning as if telling a story. Understand the theory behind machine learning and also gain experience working with different algorithms and models.
The Machine Learning course by University of Washington goes beyond the basic concepts of machine learning and explores neural networks, learning theory and vector machines. “Supervised learning” is the main focus of the class which provides the learner with correct answers at training level.
Data analysis and data scientists most commonly perform tasks like prediction and machine learning. The course offered by John Hopkins University called Practical Machine Learning – will cover basic components of building and applying prediction functions with an emphasis on practical applications. Concepts and tools you will need through the entire data science field is covered by this specialization.
Free Resources to learn Deep Learning
Deep Learning is a branch of machine learning based on linear representations of data. It attempts to model high-level abstractions in data by using multiple processing layers with complex structures. It is the buzzword in the world of neural networks and gained worldwide attention after Google’s AlphaGo defeated Lee Sedol in the game of Go last week in Korea. The AlphaGo program applied deep learning in neural networks — brain-inspired programs in which connections between layers of simulated neurons are strengthened through examples and experience. Read more about Deep Learning in this article by Nature.
Our experts at QuantInsti advise the audience to start their journey of learning Deep Learning through these two resources:
This is very good program for those who have already take a few of machine learning courses available on Udacity or Coursera and are prepared to go one step forward. You will learn how to train and optimize basic neural networks, convolutional neural networks, and long short term memory networks. This course is highly recommended by us!
Michael Nielsen’s book on Neural Networks and Deep Learning
An excellent an online free resource to get started in Deep Learning!
Recommended Books on Machine Learning
- Gareth James, Daniela Witten, Trevor Hastie and Robert Tibshirani Introduction to statistical learning ( http://www-bcf.usc.edu/~gareth/ISL/)
- Hastie, Tibshirani, and Friedman’s The Elements of Statistical Learning
- Bishop’s Pattern Recognition and Machine Learning
- David Barber’s Bayesian Reasoning and Machine Learning
- Kevin Murphy’s Machine learning: a Probabilistic Perspective
- Foundations of Machine Learning, Mehryar Mohri, Afshin Rostamizadeh, Ameet Talwalkar
- Learning From Data, Yaser S. Abu-Mostafa, Malik Magdon-Ismail, Hsuan-Tien Lin
Application of Machine Learning
While covering the fundamentals of machine learning and establishing the foundation, we also need to know how this can be implemented as a lucrative career.
A course offered by Georgia Tech called Machine Learning for Trading introduces students to real world challenges of implementing machine learning for trading strategies including algorithmic trading from information gathering to market orders. You will understand data structures used in algorithmic trading. Learn to construct software to access live equity data and assess it while making trading decisions.
For those who wants to do self-learning, this collection of ipython notebooks which are continually updated to include latest resources on popular machine learning topics are very helpful to both beginners as well as experienced data scientists.
Additional Free Resources available for further learning on MIT OpenCourseWare
A graduate/undergraduate course offered by MIT – Introduction to Convex Optimization gives its students tools and training to recognize convex optimization problems in scientific and engineering applications. It presents the basic theory and concentrating on modelling aspects and results used in applications. This course is available on MIT OpenCourseWare which means you essentially get the lecture notes of the course that was taken in Fall 2009. Once you have covered the self-learning courses, you can try assignments of this MIT course.
Prediction: Machine Learning and Statistics covers in-depth analysis of theories behind statistical learning are provided in the course while covering empirical process theory, Vapnik-Chervonenkis Theory and more.
To get first-hand experience of how machine learning and artificial intelligence is used in trading, join us for our upcoming webinar on Machine Learning.