Managerial Statistics
What We Cover
 Costis Maglaras Philip H. Geier, Jr. Associate Professor of Business |
The Managerial Statistics course offers an overview of statistical tools and methodologies that are useful in understanding and modeling uncertainty. In the context of IBS, we address two topics. The first involves ethical considerations in the collection and presentation of data. The second focuses on the use of statistical methods to test the ability to draw statistically signficant conclusions on the basis of observed data. The latter is presented through examples that illustrate the mechanics of these tools and introduce the concept of statistical significance. They also highlight the nature of conclusions that one can draw based on supporting (data-driven) evidence.
Why It Matters
How should we test whether research analysts' recommendations are biased? How much evidence do we need in order to detect racial or gender discrimination? Do doctors with financial interests in medical facilities tend to refer patients for subsequent tests and procedures more often than independent doctors? This is but a small collection of examples of issues that we face as individuals, corporations and society in our everyday life. Addressing such questions involves the analysis of data and requires a systematic (statistical) framework, which will allow us to identify in a rigorous (and defendable) fashion the set of conclusions that are supported by the data in each case. The results of such analyses may substantially impact our individual or collective decisions.
What Students Learn
Students learn the mechanics of hypothesis testing and subsequently apply these tools to a variety of real problems in which one tries to detect conflict of interest, bias or unethical behavior. Apart from serving as illustrative examples of how to use statistical tools, these cases also showcase subtleties associated with performing such analysis based on real data. This develops the students' intuition and supports critical thinking with regard to the relevance, accuracy and significance of data-driven statements and empirical observations.