Predictive Analytics and Data Modelling

Course details

Predictive Analytics and Data Modelling course dives deep into statistical methods and data techniques to predict future trends and patterns. Through hands-on exercises and real-world case studies, students learn to harness the power of data for decision-making in business scenarios.

Start Date
TBD
Target Audience
3rd & 5th semester undergraduates with foundational knowledge in data analytics. Ideal for students aiming for careers in analytics, finance, and consulting.
Duration
14-15 weeks
Learning Objectives
  • Understand the principles of predictive analytics.
  • Gain proficiency in statistical and machine learning models for prediction.
  • Acquire hands-on experience with modeling tools and software.
  • Learn to interpret and communicate modeling results effectively.
Key Features
  • Lectures by internationally recognized professors.
  • Practical labs focusing on real-world applications.
  • Group projects with peers, simulating real-world business challenges.
  • Access to industry-standard predictive analytics tools via LMS.
Learning path
  • Unit 1: Introduction to Predictive Analytics (4)
  • Unit 2: Regression Analysis (6)
  • Unit 3: Time Series Analysis (4)
  • Unit 4: Classification Algorithms (6)
  • Unit 5: Model Evaluation and Optimization (4)
  • Unit 6: Advanced Predictive Models (4)

Topics covered

  • Overview of Predictive Analytics and Its Importance
  • Types of Predictive Models
  • The Predictive Analytics Process
  • Data Collection and Preparation

  • Linear Regression
  • Multiple Regression
  • Logistic Regression
  • Model Evaluation Techniques
  • Overfitting and Regularization
  • Non-linear Regression Models

  • Basics of Time Series Analysis
  • ARIMA Models
  • Exponential Smoothing Methods
  • Forecasting with Time Series Data

  • Decision Trees
  • Random Forests
  • Support Vector Machines
  • Neural Networks
  • k-Nearest Neighbors
  • Ensemble Methods

  • Confusion Matrix and Classification Reports
  • Cross-Validation Techniques
  • Grid Search and Random Search
  • ROC and AUC Analysis

  • Gradient Boosting Machines
  • Clustering for Predictive Modelling
  • Deep Learning Foundations
  • Special Topics and Recent Trends in Predictive Modelling

Grading:

  • Participation: 10%
  • Assignments: 30%
  • Projects: 30%
  • Final Exam: 30%

Prerequisites: "Business Intelligence and Visualization" and foundational knowledge in statistics.

Reading Material: Course material available on LMS. Additional readings may be suggested by the faculty.

Faculty: A distinguished professor with expertise in Predictive Analytics and Machine Learning.

Teaching Instructors: TBD