Chevron Left
Using Machine Learning in Trading and Finance に戻る

ニューヨーク金融金融研究所 による Using Machine Learning in Trading and Finance の受講者のレビューおよびフィードバック



This course provides the foundation for developing advanced trading strategies using machine learning techniques. In this course, you’ll review the key components that are common to every trading strategy, no matter how complex. You’ll be introduced to multiple trading strategies including quantitative trading, pairs trading, and momentum trading. By the end of the course, you will be able to design basic quantitative trading strategies, build machine learning models using Keras and TensorFlow, build a pair trading strategy prediction model and back test it, and build a momentum-based trading model and back test it. To be successful in this course, you should have advanced competency in Python programming and familiarity with pertinent libraries for machine learning, such as Scikit-Learn, StatsModels, and Pandas. Experience with SQL is recommended. You should have a background in statistics (expected values and standard deviation, Gaussian distributions, higher moments, probability, linear regressions) and foundational knowledge of financial markets (equities, bonds, derivatives, market structure, hedging)....




This the best online course I've ever joined, very practical, and could be able to implement in the real world with your own thoughts plus the hints from the course.



This course was great!!! I think they skipped over a lot so it takes a lot of time to learn the details of the skills. But it definitely gives you the tools needed!


Using Machine Learning in Trading and Finance: 51 - 75 / 83 レビュー

by Karthikeyan J


good explanation of concepts and application of concepts

by Branderson J A C


Excelente curso muy interactivo y explica muy bien



Thank you for this amazing content =)




by Emma A


Thanks! - Emma

by belachkar a


Great content

by Muhammad Z H



by Rishika S



by Paolo D


I've found this course quite interesting because it makes you apply the theory explained in the lessons in the various labs both in the trading part and the Deep Learning part. Having been exposed to PyTorch previously I found Tensorflow and Keras intuitive and easy to use. The APIs are fairly similar. Even though the part about the trading strategies was treated not in-depth, I've enjoyed the labs and learned a lot from it, and got lots of inputs for future possible strategies. One downside of this course is the library used for backtesting, I wasn't able to download the data using the library, maybe because the notebooks weren't up to date with the library or the library has an undiscovered bug. Despite this little hindrance, the notebooks are really valuable!

by Matthew S


I found this course to be valuable. The tour into pairs trading, PCA, and scree plots was especially good. This course offers a guided tour of machine learning in finance, which is exactly what I sought. The Week 3 lab needs to be improved. I just completed it, and found it to be the worse lab of the two courses I have taken for this specialization. I left a full review about that lab. Unfortunately, with this lab in place, this course only deserves 4 starts.

by tsvi l


The combination of GCP+Tensorflow+Keras+Auquan+Theory is very attractive, but it is too much to cram into a single course. I thought it would be much deeper on the financial theory side. The Auquan part is directly useful as one needs a good backtesting tool and this is probably the best/only one. But I do not necessarily need Keras/TF and I definitely do not need GCP details so much, as I may end up using a different cloud and ML library.

by Umendra C


There are some good trading strategy concepts introduced in this course. However, the link between trading strategy and ML is superficial. Maybe, if the course part related to ML and GCP is taken out, this course would be a sure 5-start worth.

by Andrey P


It's not deep enough to understand how to implement ML in algorithmic trading, but the course explains some helpful concepts like pairs trading and Kalmar filter.

by Gustav K F Y


Although the often glitches in the Google Cloud platform prevented me to complete the exercises, the course material is very useful.

by David C C R


Useful for people who have previous knowledge of coding and trading basics. I get a lot of ideas from this course. I will recommend.



The concepts and algorithms are great. Unfortunately, the last 4 Jupyter Notebooks of the course did not work !!!!!

by Oleksii K


If that unrelated Google Cloud part were thrown away, it would be a decent course.

by Siro G


I wish better examples to cover everything was said during the lectures

by Mohak S


Labs should be more engaging. And should probably move to Colab.

by Justin C


Useful but some codes are outdated and cannot be run.

by Kevin S


Some resources are outdated

by Hongkai Y


The topic is interesting, but the content introduced in this course is rather shallow. One could grasp the basic idea of pair trading / momentum trading after the course, but will still be very struggled to implement one. Also, the last two labs of the course couldn't work. For some reason the program cannot install auquan toolbox thus the programs don't work, which is very frustrating.

by Nitin K


One of the instructor from Google Cloud just reads the slide instead of giving much insight.

by David G


This quite a low effort course. Some labs are just copy/pastes of blog posts. The "QwikLabs" are slow and painful to use, and most of them don't work anymore because they use Python packages that haven't been updated in many years.

And at no point do they actually demonstrate a strategy that shows any promise in the real world. Just vague talk of "hedging market risk" without evidence of anything performing better than chance. I have a funny feeling this is all an exercise in being fooled by randomness.

by Chris C


I understand that this course is not about building the world's best high frequency trading model. On the other hand, it should at least be about making a serious attempt to build a trading model, which is something the series of courses has so far neglected to even attempt.