High-Level Interface Between strongRstrong and Excel.pdfVIP

New URL: http://www.R-/conferences/DSC-2003/ Proceedings of the 3rd International Workshop on Distributed Statistical Computing (DSC 2003) March 20–22, Vienna, Austria ISSN 1609-395X Kurt Hornik, Friedrich Leisch Achim Zeileis (eds.) http://www.ci.tuwien.ac.at/Conferences/DSC-2003/ High-Level Interface Between R and Excel Thomas Baier Erich Neuwirth Abstract Spreadsheet programs in general and Microsoft Excel in particular imple- ment a very special paradigm of computing: automatic creation of depen- dency structures for calculation, and automatic recalculation when input val- ues change. This is rather different from the usual model of batch processing underlying most statistics software, e.g. R. Since some time now, a COM server implementation for R has been avail- able. As a more recent development, R can also act as COM client, using Excel as a server. So we can either think of R as an extension of Excel, or of Excel as a utility program (and data editor) for R. We will discuss the differences between these 2 approaches, and their rela- tive merits under different circumstances. This will be done by showing exam- ples of applications implementing these rather different models for statistical computation. Emphasis will be more on the philosophy and style of statistical analysis than on technical implementation details. 1 Introduction r R (Ihaka and Gentleman, 1996) and Microsoft Excel (Microsoft Corporation (1999)) have