Module Code - Title:
FI6015
-
DEEP LEARNING FOR FINANCE
Year Last Offered:
2025/6
Hours Per Week:
Grading Type:
N
Prerequisite Modules:
Rationale and Purpose of the Module:
This module is a core module for the MSc in Machine Learning For Finance and builds and extends on the MSc in Machine Learning for Finance. The aim of this module is to provide candidates with practical, practitioner-oriented skills in deep learning algorithms. Deep Learning is a subfield of machine learning concerned with algorithms inspired by the structure and function of the brain called artificial neural networks. Specifically, this module uses recurrent neural network (RNNs), a class of artificial neural networks, where connections between nodes form a directed graph along a temporal sequence. This allows it to exhibit temporal dynamic behaviour that is typical of asset price fluctuations. This module will delve into Long Short Term Memory networks - (LSTMs), a particular form of RNNs that are especially useful in financial time series analysis. The module will culminate in a detailed, real-life, application of LSTMs using contemporary information from global stock markets.
Syllabus:
The module will start with a revision of ANNs and address specific the particular issues faced by data from financial markets. Data normalization and processing sequential data will be discussed and various mechanisms used to overcome these problems will be trialed. The module will delve into the mechanics of RNNs and, in particular, LSTMs. Contemporary stock market data will be used to train and test the model. A good deal of consideration will be given to model parameters such as the learning rate, in order to minimize the loss function.
Learning Outcomes:
Cognitive (Knowledge, Understanding, Application, Analysis, Evaluation, Synthesis)
At the end of this module students will be able to: 1. Demonstrate an understanding of the problem of Long-Term Dependencies in RNNs
2. Demonstrate an understanding of how LSTMs work in practice on stock price data
3. Describe the use of training and testing sequential data
4. Know how to interpret the results of a LSTM
5. Demonstrate a working model of LSTM using stock price data
6. Identify data characteristics that assist/impede model fitting
7. Illustrate the efficacy of the model using visual tools
Affective (Attitudes and Values)
At the end of this module students will be able to: 1. Demonstrate an appreciation of specialised modelling techniques.
2. Acknowledge the model assumptions and limitations.
Psychomotor (Physical Skills)
N/A
How the Module will be Taught and what will be the Learning Experiences of the Students:
The module will be delivered using online lectures, virtual labs and tutorials.
Research Findings Incorporated in to the Syllabus (If Relevant):
Prime Texts:
McNelis, P.D. (2005)
Neural networks in finance: gaining predictive edge in the market
, Academic Press.
Other Relevant Texts:
Heaton, J.B., Polson, N.G. and Witte, J.H. (2017)
Deep learning for finance: deep portfolios. Applied Stochastic Models in Business and Industry, 33(1), pp.3-12.
,
Sirignano, J. and Cont, R. (2019)
Universal features of price formation in financial markets: perspectives from deep learning. Quantitative Finance, 19(9), pp.1449-1459
,
Deng, Y., Bao, F., Kong, Y., Ren, Z. and Dai, Q. (2016)
Deep direct reinforcement learning for financial signal representation and trading. IEEE transactions on neural networks and learning systems, 28(3), pp.653-664.
,
Programme(s) in which this Module is Offered:
Semester(s) Module is Offered:
Spring
Module Leader:
darren.shannon@ul.ie