Predictive Analytics
5 – 6 hours per week
Course Information
Course Length : 1 week of pre-requisites and 6 weeks of content
Estimated Effort : 5 – 6 hrs per week
Course Overview
Decision-makers often struggle with questions such as “What should be the right price for a product?,” “Which customer is likely to default in their loan repayment?,” “Which products should be recommended to an existing customer?” and so on. The list of questions is endless and finding right answers to these questions can be challenging.
In order to make the task of decision-making simpler, Predictive Analytics aims to predict the probability of the occurrence of a future event such as customer churn, loan defaults, and stock market fluctuations, thus enabling effective business management. Models such as multiple linear regression, logistic regression, auto-regressive integrated moving average (ARIMA), decision trees, and neural networks are frequently used in solving predictive analytics problems. Regression models help us to understand the relationships among these variables and how their relationships can be exploited to make decisions. Hence, Predictive Analytics is now emerging as a competitive strategy across many business sectors and can set apart high performing companies.
This course taught by Prof Dinesh Kumar is designed to prepare the learner for a career in the field of data analytics; and is also suitable for students/practitioners interested in improving their knowledge in the field of predictive analytics. So, if you are in the quest for the right competitive strategy to make companies successful, then join this course to master the tools of predictive analytics.
Dinesh Kumar
Professor
Decision Sciences & Information Systems
Indian Institute of Management Bangalore (IIMB)
Course Syllabus
Week 0: Course Prerequisites
- Course Overview
- Descriptive Analytics
- Probability Distribution
- Hypothesis Testing
- Analysis of Variance
- Software Installation
Week 1: Introduction to Analytics
- Introduction to Analytics
- Analytics in Decision Making
- Game changers & Innovators
- Predictive Analytics
- Experts view on Analytics
Week 2: Simple Linear Regression (SLR)
- Case-let Overview
- Introduction to Regression
- Model Development
- Model Validation
- Demo using Excel & SPSS
Week 3: Multiple Linear Regression (MLR)
- Multiple Linear Regression
- Estimation of Regression Parameters
- Model Diagnostics
- Dummy, Derived & Interaction Variables
- Multi-collinearity
- Model Deployment
- Demo using SPSS
Week 4: Logistic Regression
- Discrete choice models
- Logistic Regression
- MLE Estimation of Parameters
- Logistic Model Interpretation
- Logistic Model Diagnostics
- Logistic Model Deployment
- Demo using SPSS
Week 5: Decision Trees and Unstructured data analysis
- Introduction to Decision Trees
- CHI-Square Automatic Interaction Detectors (CHAID)
- Classification and Regression Tree (CART)
- Analysis of Unstructured data
- Naive Bayes Classification
- Demo using SPSS
Week 6: Forecasting and Time series Analysis
- Forecasting
- Time Series Analysis
- Additive & Multiplicative models
- Exponential smoothing techniques
- Forecasting Accuracy
- Auto-regressive and Moving average models
- Demo using SPSS
After completing this course, learners should be able to:
-
Apply predictive analytics tools to analyse real-life business problems.
- Demonstrate case-based practical problems using predictive analytics techniques to interpret model outputs.
- Examine regression, logistic regression, and forecasting using software tools such as MS Excel, SPSS, and SAS
Students pursuing post-graduate program, Researchers, Academicians and Researchers
This course is very good for beginners. For those who dont have statistical background, this course covers everything from scratch. I really like they in which prof. Dinesh Kumar structured the curriculum. Obviously it is impossible to cover everything in one video lesson but this course is good to start with.
(Ankita P)
The content is well structured and explained. Found it very useful being a beginner in this area. The practical implementation explained in the course (Apollo Hospital, L&T , Die Another Day hospitals) helped to relate to the theory. The demo of SPSS was also useful.
(Anoop Kumar Chenayil)
Great course! I have taken 2 courses in stats in college before but did not learn nearly as much as I did from this single course. The topics were really interesting and useful with respect to today’s job market. The subjects are practically presented with lots of examples without too much unnecessary theoretical stuff. Of course, for someone like me with a limited background in stats it was difficult to grasp a lot of the material presented,but with some additional research, mostly on Youtube, I was able to keep up with the course. I am looking forward to seeing more from Prof Kumar!
(Ivan Tuhchiev)
This course is really cool. Solid handful of knowledge and it’s applications. Sometimes too many immediate conclusions, and it is necessary to check different sources to make the topic clear. However this is really solid and thorough piece of course. Thanks a lot.
(Jakub S)
This is my maiden experience with MOOC. Predictive analysis course was highly informative and the case studies was really useful from the application point of view. Prof. Dineshs’ lectures are highly impressive. Kudo’s Prof. Looking forward to more from you.
(Nambi ST)