Introduction to Fundamentals of Machine Learning in Finance

Loading...

En provenance du cours de New York University Tandon School of Engineering

Fundamentals of Machine Learning in Finance

111 notes

Explore Related Videos

At Coursera, you will find the best lectures in the world. Here are some of our personalized recommendations for you

New York University Tandon School of Engineering

Fundamentals of Machine Learning in Finance

111 notes

Cours 2 sur 4 dans Specialization Machine Learning and Reinforcement Learning in Finance

The course aims at helping students to be able to solve practical ML-amenable problems that they may encounter in real life that include: (1) understanding where the problem one faces lands on a general landscape of available ML methods, (2) understanding which particular ML approach(es) would be most appropriate for resolving the problem, and (3) ability to successfully implement a solution, and assess its performance. A learner with some or no previous knowledge of Machine Learning (ML) will get to know main algorithms of Supervised and Unsupervised Learning, and Reinforcement Learning, and will be able to use ML open source Python packages to design, test, and implement ML algorithms in Finance. Fundamentals of Machine Learning in Finance will provide more at-depth view of supervised, unsupervised, and reinforcement learning, and end up in a project on using unsupervised learning for implementing a simple portfolio trading strategy. The course is designed for three categories of students: Practitioners working at financial institutions such as banks, asset management firms or hedge funds Individuals interested in applications of ML for personal day trading Current full-time students pursuing a degree in Finance, Statistics, Computer Science, Mathematics, Physics, Engineering or other related disciplines who want to learn about practical applications of ML in Finance Experience with Python (including numpy, pandas, and IPython/Jupyter notebooks), linear algebra, basic probability theory and basic calculus is necessary to complete assignments in this course.

À partir de la leçon

Fundamentals of Supervised Learning in Finance

What is Machine Learning in Finance?5:37

Introduction to Fundamentals of Machine Learning in Finance4:40

Rencontrer les enseignants

Igor Halperin

Welcome to Fundamentals of Machine Learning in Finance.

You have just completed our guided tour course that is

called Guided Tour of Machine Learning in Finance.

Now, this course will continue what we started in the first course.

But this time, we go a bit deeper and consider

different types of machine learning in finance sequentially and in more details.

We'll start exactly where we left it in

the guided tour course, namely supervised learning.

In the guided tour course,

we talked about a lot about supervised learning.

We started with Linear Regression as a machine learning algorithm and saw how it works,

and how it can be seen as a special case of Non-linear Regression.

Then we saw how both can be implemented in Python packages such as scikit-learn,

starts morals, and tons of law.

Then we talked about how a Non-linear Regression can be implemented with Neural Networks.

After that, we talked about Binary Classification and introduce

Logistic Regression as one particular approach to binary classification.

We also saw how simply a Logistic Regression and

Binary Classification can be implemented with Neural Networks.

And then, we tried them with the FDIC data.

So, what we did in this guided tour course,

we started with simplest models of main tasks of supervised learning,

namely regression and classification,

and showed how they can be addressed using either very simple models

such as linear regression or highly complex model such as molecule or neural networks.

As I mentioned in the guided tour course,

neural networks are very flexible and highly scalable.

They can be used to solve any problems of regression or classification.

Why then to study other methods?

Well, there are many answers to such very legitimate question.

The first answer is suggested by

the no free lunch theorem that we discussed in the first course.

To remind you it's statement,

it says that there is no single model that is

universally better than any other model across all the means.

In particular, it means that if neural networks beat other models

on visual or text data when there is lots of data,

it doesn't mean yet that they will get magically best models for finance.

But in finance, data is qualitatively different from visual or text data.

In particular, typically we do not have lots of data in finance.

Moreover, these data are typically noisy and often non-stationary.

So, it will be good idea in this case to get familiar with

other methods of supervised learning that are not called neural nets.

And this is what we are going to do in this course.

Even though we will not forget about neural networks.

So, here is what we will do in this course,

in the first week,

we will talk more about supervised learning.

More specifically, we will talk about two other methods of

supervised learning that we did not cover in the first course.

The first type of algorithms is called Support Vector Machines.

The second type is an approach based on decision trees and their extensions.

Both these methods are widely used by machine learning

researchers and sometimes they work better than a neural networks for financial data.

Why choose them over neural nets?

Because of their simplicity or better interpretability.

The second week of this course course will be devoted to unsupervised learning.

Again, as in the first course,

we will start with simplest,

and then gradually build it up.

We will start with

the Principal Component Analysis as a non-supervised dimension reduction method.

And after that, we will talk about

non-linear neural methods for dimension reduction such as [inaudible] order.

In week three, we will continue with topics of unsupervised learning.

We will talk about clustering methods,

and data visualization methods such as [inaudible].

And finally, in week four,

we will talk about modeling sequence and time series data.

We will also talk in this week about reinforcement learning.

We will introduce basic concepts of reinforcement learning

that will serve to give a short overview of this topic,

and an introduction to our next course in this specialization.

Finally, you will have a course project where you will

apply methods of unsupervised learning to stock analogies.

So, good luck with the course and let's get it started.

Explorer notre catalogue

Rejoignez-nous gratuitement et obtenez des recommendations, des mises à jour et des offres personnalisées.

Coursera propose un accès universel à la meilleure formation au monde, en partenariat avec des universités et des organisations du plus haut niveau, pour proposer des cours en ligne.

© 2019 Coursera Inc. Tous droits réservés.

Télécharger dans l'App Store

Disponible sur Google Play