Course description
Modern world organisations require professionals and teams that can perfectly use complex and huge volumes of datasets and make informed choices by deriving great strategic insight from these datasets. Machine and deep learning have become a very important part of the decision-making process in the financial sector. It can be utilised to assess risks, streamline operations, discern choices, inform investment decisions, and design actionable plans. Machine learning skills can improve the performance of a financial professional dramatically.
Which areas can be improved by machine learning in the financial sector?
Machine learning has the potential to improve several areas in the financial sector, including:
- Fraud detection and prevention
- Credit scoring and risk management
- Customer service and support
- Algorithmic trading
- Portfolio optimisation and asset management
- Predictive modelling for financial forecasting
- Anti-money laundering
- Insurance claims processing
Upcoming start dates
Suitability - Who should attend?
Who should attend?
The Machine Learning and Deep Learning in Finance and Investments Training Course by Rcademy is ideal for the following professionals:
- Data scientists
- Investment managers
- Portfolio managers
- Financial engineering
- Risk managers
- Analysts in Finance
- Financial econometrics scientists
- Statisticians
- Financial managers
Outcome / Qualification etc.
The objectives of The Machine Learning and Deep Learning in Finance and Investments Training Course by Rcademy are to enable participants to:
- Learn how to use modern and cutting-edge research in machine learning to make better models
- Gain practical experience in creating predictive models using decision trees, neural networks, support vector classifiers, activation layer, and regression algorithm
- Discover ways to implement, describe, and list the main differences between the working of the DBSCAN clustering algorithm and k-means clustering
- Understand how and why predictive models fail and how they can be improved by applying gradient boosting, hyper-parameter tuning, cross-validation, and many other techniques
- Apply the learnt skills and knowledge to develop predictive models used in live trading and understand how the models are used for live trading
- Masterartificial intelligence techniquesand packages critical for financial markets prediction
- Understand how to build supervised and unsupervised models
- Use classical machine learning techniques
Training Course Content
Module 1: Cross-Sectional Data and Machine Learning
- Fraud detectionusing deep learning and machine learning
- Cryptocurrencies
- Big data and computations in finance
- Prediction entropy
- Robo-advisors in Fintech
- Machine learning and predictions using neural networks
- Statistical modelling vs machine learning
Module 2: Probabilistic Machine Learning Modelling
- Frequentist inference from data
- Sequential Bayesian updates
- Bayesian vs Frequentist estimations
- Predictions using Bayesian updates
- The Beta distribution
- Model selection process
- Occam’s Razor
- Bayesian inference
- The Bias-variance tradeoff for estimators
- Model averaging
- Model selection
- How do select models from many models
- Maximum likelihood estimation
- Online learning prediction
- Hidden indicator variable representation of mixture models
Module 3: Introduction to Machine Learning and Deep Learning
- Using machine learning as opposed to using statistics
- Machine learning applications in predicting risks, credit risk, portfolio optimisation and key selection
- Deep learning methods
- Reinforcement learning
- Supervised vs unsupervised learning
- Data vendors and their contribution to financial machine learning
- Fintech in machine learning
- Alternative data
- Big data
- Fintech
Module 4: Supervised and Unsupervised Learning
- Cross-sectional data
- Evaluating machinelearning algorithms
- Time series analysis
- Hierarchical clustering
- Clustering techniques
- Affinity propagation
- K-means
- Regression, neural networks
- Distance measurement
- Random forest
Module 5: Decision and Random Trees
- Introduction
- Regression trees
- Forecasting bond returns using macroeconomic variables
- Classification trees
- Default prediction based on accountancy data
- Issues common to classification and regression trees
Module 6: Sequential Learning
- Introduction
- Exponential smoothing
- Stability in autoregressive processes
- Stationarity
- Partial autocorrelations
- Autoregressive processes
- Heteroscedasticity
- Predicting events
- Principal component projection
- Convexity and inequality constraints
- Back-propagation
- Computational considerations
- Composition with ReLU activation
- Function approximation with deep learning
Module 7: Gaussian Processes Bayesian Regression
- Computational properties of Gaussian and Bayesian regression
- Gaussian processes in finance
- Structure exploiting inference
- Mesh-free GPs
- Exercise
- Case study and practical example
Module 8: Feedforward Neural Networks
- Introduction
- Probabilistic reasoning
- VC dimension
- Approximating with compositions of functions
- Function approximation with deep learning
- Geometric interpretation of feedforward networks
- Training, testing, and validation
- Model averaging via dropout
- RNN memory using partial autocovariance
- Generalised recurrent neural networks
- Neural network exponential smoothing
- ∝-RNNs
- Long short-term memory
- Convolutional neural networks
- Autoencoders
Request info
Rcademy
Rcademy is a global training and consultation organisation set out to bridge the gap between you now and what you can be in the near future. We are facilitators of knowledge impartation. Our team of established and experienced training enthusiasts...