Presented to the monthly Delaware ASA meeting in February 2010.
Abstract:
For all its advantages & disadvantages, Microsoft Excel is an ubiquitous part of statistics today. Since it is limited for complex calculations, statisticians rarely do statistical analyses using Excel. However, it is a standard for data collection and presentation, and provides a common way of communication between statisticians and non-statisticians. Although simulations usually require large numbers of simple calculations, and Excel allows iterative calculations, simulation in Excel usually requires add-in programs, or programming in Visual Basic. However, a few simple programming tricks allow it to be used for simulations without any additional programming, and a few statistical functions programmed in VBA make it useful for a variety of common problems. While Excel simulations can be slow, they inherit many advantages from Excel: ease of coding, debugging tools, graphics, and being understandable by non-statisticians. All VBA code and spreadsheet examples will be posted on the Delaware ASA website.
Biography:
Dennis Sweitzer, PhD is a Principal Statistician at AstraZeneca supporting projects in Clinical Development in the area of Neuroscience. He received his PhD from the University of Delaware Mathematics Department. At AstraZeneca he has been involved in statistical design, analysis and interpretation of clinical studies as well as providing modeling and simulation development to support study management and decision making.