Syllabus and course description:
This course is designed for students with a good background and understanding in statistics. It is also offered to 3rd or 4th year college students that intend to enroll into a quantitative graduate programs. See prerequisites below.
This course is offered at a time when demand for data-analytic consulting is almost unprecedented. Because of its relevance to a number of disciplines, it is an elective course in entrepreneurship, marketing and statistics (besides being a substitute for core courses in statistics).
The course presents a unique framework of consulting analytics that focuses on one of the most abundant kind of variables in the business world – categorical variables.
Because of its relevance to a number of disciplines Individuals measured on these variables are classified into categories. Decision to buy, Gender, Region of residence, attitudinal variables (such as satisfaction, agreement, loyalty, life-style, political orientation) are just a few examples of such variables. As it turns out, especially in the current business Big Data era, such variables have an impressive predictive power for describing consumer behavior. Data analytics of such variables requires a specific methodology that this course extensively provides. More information is given in the course description below.
Course description:
You decide to establish a start-up in business consulting. You search the Internet and find to your dismay well over 650 companies in that area, each one claiming to be best and unique. In order to compete in this arena you need to have the ability to identify upcoming trends and new problems in the business world AND to be able to provide original, innovative and applicable solutions to these problems. One such example that is not dealt by many of the existing consulting companies is the following shelf-planning problem.
Imagine a customer walking into a deli store on a Sunday morning intending to buy bagels. There are only two bagels on the shelf. What would you predict the person would do? Hurry up and buy the only remaining bagels before they are gone? Would she consider the two remaining bagels as being the least fresh, touched and left by all former customers, and therefore decide to wait for a fresher batch? As a consultant to the store manager, how would you determine the minimum number of bagels that should be on the shelf at a given time in order to avoid making customers reluctant to buy?
As is shown in the course, the specific methodology that solves the above-mentioned problem, can also be used for many consulting problems in totally different areas.
Unlike research, consulting is a problem-solving endeavor that requires a great deal of resourcefulness and is fueled by experience. This course is meant to give future consultants and entrepreneurs important and relevant tools for dealing with powerful, insightful and ethical consulting in the business world.
The vast (and successful) consulting experiences of faculty were collected, selected, and are shared with the students in this application-oriented course.
As is frequently happens in consulting, the client imposes constraints that are not addressed by the available optimal scientific solution. Resorting to sub-optimal solutions is almost never taught, yet, it is of top importance in practical quantitative consulting. This course accommodates client-imposed constraints to provide sub-optimal solutions, compares them the optimal solution, and presents such comparisons to the (surprised and appreciative) client.
Among other important consulting problems, this course addresses issues
• Analyzing customer attrition as a process (rather than as an event-driven phenomenon).
• Optimal inventory management.
• Prediction of a customer's purchase behavior (buying intentions, buying propensity, etc.) from the customer's patterns of usage of media, life style, political orientation, etc.
• Analysis of satisfaction -how to create a VALID satisfaction scale, how to rank products by satisfaction of customers, how to detect easy-to-please customers, etc.
• Analysis of brand loyalty -how to measure loyalty, how to determine whether loyalty to certain brands exists, and how to quantify it.
• How to systematically obtain brand imagery from consumer's data.
The course is taught in a way that emphasizes the interpretation of results rather than computations. To aid in the analysis, an interactive and user-friendly, R-based software made up by the instructor, will be used. This easy-to-use program completely eliminates the need to know programming ahead of the course. Students don’t waste time on computation, but time is dedicated to understanding the proposed solutions and how to interpret the results of the analyses to clients. The teaching style promotes understanding rather than
memorizing. Although this course uses statistical reasoning, it is NOT mathematical in nature.
The course offers insightful understanding of quantitative uncertainty that prevails so abundantly in the business world.
This course may well prepare you for statistically advanced courses such as data mining, big data and machine