Master of Management Studies (M.M.S) Predictive Analytics Syllabus - Mumbai University 2026

Course objectives

• 1 To introduce foundational concepts, tools, and techniques of predictive analytics.
• 2 To equip students with practical skills in data modeling, forecasting, and statistical analysis.
• 3 To develop analytical abilities for applying predictive analytics in decision-making across various business functions.
• 4 To foster understanding of predictive models such as regression, classification, and time-series forecasting.
• 5 To cultivate insights into ethical considerations and challenges associated with predictive analytics implementation.

Course outcomes

• CO1: The basic concepts of predictive analytics, including types of analytics and their applications in business decision-making.
• CO2: The use of regression models for prediction and estimation in real-world business scenarios.
• CO3: Logistic regression and classification techniques to solve business problems, including customer segmentation and event prediction.
• CO4: Data using forecasting methods to make accurate predictions based on business data.
• CO5: The performance of predictive models and assess their relevance to business outcomes.
• CO6: Actionable insights and predictive models using software tools to address real-world business challenges

Unit 1 CO1, CO2 · 3 hours

Introduction to Predictive Analytics and Types of Analytics - Types of analytics: Descriptive, Predictive, and Prescriptive. - Real-life examples of predictive analytics in various industries (finance, marketing, healthcare). - Benefits of using analytics for decision-making.

Unit 2 CO1, CO2 · 3 hours

Understanding Analytics in Business Decision- Making - How predictive analytics helps in forecasting future events and behaviours. - Case studies of successful businesses using predictive analytics.

Unit 3 CO2, CO3 · 3 hours

Regression Analysis: Simple vs. Multiple Linear Regression - Explanation of Simple Linear Regression: Concept, assumptions, and applications. - Moving to Multiple Linear Regression (MLR): How multiple variables influence the dependent variable. - Model fitting, residuals, and interpretation of coefficients. - Hands-on practice: Fitting a regression model using sample data.

Unit 4 CO3, CO4 · 3 hours

Applying Multiple Linear Regression to Predict Business Outcomes - Identifying and handling multicollinearity, outliers, and model diagnostics. - Assumptions of MLR and how to check for violations. - Interpreting model outputs and making data- driven decisions. - Hands-on practice: Solving a real-world problem using MLR.

Unit 5 CO3, CO4 · 3 hours

Introduction to Logistic Regression and Classification Problems - Introduction to classification problems and the need for logistic regression. - Difference between continuous and categorical outcomes. - Concept of odds ratio and logit function in logistic regression. - Applications of logistic regression (e.g., customer churn, fraud detection). - Hands-on practice: Building a logistic regression model for classification.

Unit 6 CO4, CO5 · 3 hours

Using Logistic Regression for Customer Segmentation and Event Prediction - Using logistic regression for binary classification (e.g., predicting customer churn). - Multinomial logistic regression for multi-class classification. - Building a customer segmentation model based on demographic and behavioral data. - Evaluating model performance using metrics like Accuracy, AUC, Precision, and Recall. - Hands-on practice: Solving a customer prediction problem using logistic regression.

Unit 7 CO4, CO5 · 3 hours

Decision Trees and Classification Methods - Introduction to decision trees: Concept, structure, and splitting criteria (e.g., Gini index, entropy). - Building a decision tree for classification. - Pruning and avoiding overfitting in decision trees. - Comparison with other classification methods like SVMs and k-NN. - Hands-on practice: Creating a decision tree model for classifying customers.

Unit 8 CO2, CO4 · 3 hours

Forecasting Methods and Time Series Analysis - Introduction to time series data and its components: Trend, Seasonality, Noise. - Forecasting methods: ARIMA, Exponential Smoothing, and Moving Averages. - Evaluation of forecast accuracy: Mean Absolute Error (MAE), Root Mean Squared Error (RMSE). - Case study: Predicting sales or demand for a business using time series methods. - Hands-on practice: Applying ARIMA for sales forecasting.

Unit 9 CO5, CO6 · 3 hours

Analyzing Structured vs. Unstructured Data, Sentiment Analysis - Difference between structured and unstructured data (e.g., text, images, social media). - Introduction to sentiment analysis and text mining. - Using natural language processing (NLP) techniques for sentiment analysis. - Hands-on practice: Creating word cloud

Unit 10 CO3, CO5, CO6 · 3 hours

Hands-on Practical Applications and Case Studies in Predictive Analytics - Review of key concepts learned: Regression, Logistic Regression, Decision Trees, Forecasting. - Real-world business case studies: Predicting customer behaviour, sales forecasting, and market trends. - Group activity: Working on a predictive analytics case study in a team, applying multiple techniques. - Final project work: Developing a predictive

Textbooks

• 1 Essentials of Business Analytics - Descriptive, Predictive and Prescriptive Analytics, Jeffrey D. Camm, James J. Cochran, Michael J. Fry, Jeffrey W. Ohlmann, David R. Anderson, Cengage Learning, 2nd Edition, 2017
• 2 Business Analytics, Albright & Winston, Cengage Learning, 5th Edition
• 3 Data Science for Business: What you need to know about data mining and data- analytic thinking, Provost, F., & Fawcett, T., O`Reilly Media, Inc., 2013.

Reference Books

• 1 Applied Simulation Modeling: Andrew F. Seila, V. Ceric and P. Tadikamalia,Cengage Learning, 2004
• 2 Decision Making under Uncertainty with RISK Optimizer: Wayne Winston,Kelley School of Business, Indian University, Palisade Corporation, 2010.
• 3 Data Mining for Business Analytics: Concepts, techniques, and applications inR. Shmueli, G., Bruce, P. C., Yahav, I., Patel, N. R., & Lichtendahl Jr, K. C.,John Wiley & Sons., 2017
• 4 Uncertainty & Risk Analysis - Chris Rodger and Jason Petch, Business Dynamics, PricewaterhouseCoopers United Kingdom firm, 1999.