Course

Unit 1

Introduction to Predictive Modeling
Introduction – Meaning – Models – Types of Models – Process of predictive modelling Tidying Data and Measuring Performance – Tidying data – Categorizing data quality – Performance metrics – Cross Validation – Curves.

Unit 2

Linear Regression
Linear regression – Simple linear regression – multiple linear regression – assessing linear regression models – problems with linear regression – feature selection – regularization – polynomial regression
Generalized Linear Models – classifying with linear regression – logistic regression – assessing logistic regression – regularization with the lasso – classification metrics – extensions of the binary logistic classifier – Poisson regression – negative binomial regression.

Unit 3

Neural Networks and Support Vector Machines
Neural Networks – the biological neuron – the artificial neuron – stochastic gradient descent – multilayer perceptron networks – the backpropagation algorithm – radial basis function networks
Support Vector Machines – maximal margin classification – support vector classification – kernels and support vector machines – multiclass classification with support vector machines.

Unit 4

Tree-Based Methods and Probabilistic Graphical Models
Tree-Based methods – the intuition for tree models – algorithms for training decision tress –
improvements to the M5 model – Dimensionality reduction. Ensemble methods – Bagging – Boosting.
Probabilistic Graphical models – little graph theory – Bayes’s theorem – conditional independence – Bayesian networks – the naïve Bayes classifier.
Topic modeling – an overview of topic modeling – Latent Dirichlet Allocation (LDA).

Unit 5

Recommendation Systems and Deep Learning
Recommendation Systems – Rating matrix – collaborative filtering – singular value decomposition.
Deep Learning – Scaling Up project – characteristics of big data – training models at scale –
Introduction to Deep Learning – Machine Learning or Deep Learning – Deep Learning Models.

Program Outcomes:
PO1: Enriched knowledge with new ideas and techniques essential for business and management.
PO2: Mastery of specific skills in business.
PO3: Capability to acquire and handle any position in the business.
PO4: Develop analytical interpretative, and presentation skills regarding research in commerce and management.
PO5: Acquaintance with recent trends in commerce and management.

About The Course

1. This course helps students understand predictive analytics and application techniques used in the finance domain.

Course objectives

CO1: To understand the significant methods of predictive modeling beyond the black-box thinking

CO2: To gain knowledge of data modeling and model-performing metrics

CO3: To gain knowledge of neural networks and applications in the finance domain.

CO4: To understand the training steps and test the predictive models.

1. James D. Miller, Rui Miguel Forte – Mastering Predictive Analytics with R, 2nd edition – Packt Publishing
2. Jeffrey Strickland – Predictive Analytics using R – Lulu.com publication
3. Rui Miguel Forte – Mastering Predictive Analytics with R – Packt Publishing