GRADUATE AND EXECUTIVE CERTIFICATE PROGRAMS

Statistical Modeling I

This session provides an introduction to the statistical methods used in decision modeling, including: (1) review of the mean, median, mode, standard deviation, normal curve, graphs and charts; (2) sampling - why sample, uses of sampling, comparison of sample survey methods on 10 dimensions, sample design and choosing random samples; (3) making estimates from samples - central limit theorem, confidence intervals for means and proportions; and (4) computing sample size for estimating means and proportions.

Statistical Modeling II

Once we are familiar with the basic concepts of descriptive statistics we can take on more complex tasks. In this session students investigate how to implement these concepts in Excel and how to use statistics to generate conclusions. The first topic covered here is hypothesis tests; the primary decision-making tool in statistics. We begin by applying hypothesis tests to one-variable problems. More realistic models require multiple variables. In working with multiple variables we build regression models. The objective is to develop a model where the dependent variable is explained by one or more independent variables. This allows identification of the factors that determine the outcome of the dependent variable. Students will become familiar with implementing regression in Excel and interpreting the Excel output.

Decision Modeling I

In this class meeting, the focus is on decision-making under uncertainty.  For a large class of decision problems, the outcome of a decision is uncertain, but the decision-maker is able to list the possible outcomes of a decision and assign a probability to each of them. Sources of uncertainty include such factors as consumer demand, competitors' behavior, and acts of nature. To address these uncertainties in a decision-making context, decision trees have proved to be a powerful graphical tool.  Decision trees allow a decision-maker to consider the possible outcomes of his decisions in a systematic way and to draw correct conclusions with respect to the best course of action. Spreadsheet implementation of decision trees further extend to the decision-maker the ability to perform sensitivity analysis to gain a deeper understanding of how sensitive the optimal solution is to changes or inaccuracies in model assumptions.

Decision Modeling II

Linear programming (LP) is a tool for the mathematical optimization of complex decision problems.  This session begins with a fun, hands-on, interactive problem-solving exercise drawn from production planning.  This simple example is used to introduce the basic concepts behind LP as a simple mathematical model of the exercise is developed and then solved using the Solver in Microsoft's Excel.  The session concludes with students building small models in Excel using examples drawn from distribution planning and financial management.

Decision Modeling III

Some real world problems cannot be modeled using pure optimization techniques, particularly those that involve an element of chance. In these situations we often resort to simulation, a methodology that combines decisions (e.g., how many service representatives to hire) and uncertainties (e.g., the demand for service). Simulation involves building a computer model that generates probabilistic outcomes based on the decisions we make and the uncertainties we face. One can use simulation models to determine how specific decisions will perform in practice without experiencing their effects firsthand. In this lecture, we will demonstrate how many complicated problems can be analyzed using simulation models built in spreadsheets.

Revenue Management

This lecture is designed to introduce you to some of the models and methods used in the emerging field of revenue management (RM). The problem that motivates RM is ancient: how does a company manage its capacity and/or prices to extract the greatest possible revenue from the marketplace? The RM approach is to exploit differences in customer segments and their willingness to pay. For example, a person who books a room in an upscale hotel three days in advance is typically willing to pay much more than someone booking the same room three months in advance. RM focuses on how to manage these different types of customer demands to maximize revenues.