----------------------------------------------------------- The Boston Area SAS Users Group Message List - November 5, 1998 ----------------------------------------------------------- Contents: - Quarterly Meeting, December 10, 1998 - BASUG Web Site - BASUG Contacts - How to subscribe/unsubscribe from this Message List ----------------------------------------------------------- ----------------------------------------------------------- Meeting Announcement ----------------------------------------------------------- ----------------------------------------------------------- WHAT: Quarterly BASUG Meeting WHEN: Thursday, December 10, 1998 - 8:30AM to 12:30PM WHERE: Epsilon Data Management 50 Cambridge Street Burlington, MA (directions below) AGENDA: - "SAS/Insight--Statistical Application Demo" - "Nonlinear Least Squares Analysis of HIV Viral Load Data" - "An Introduction to Mixed Models in SAS" CONTACT: Ming Yuan Tel: (800) 225-1970 (x6987) E-mail: myuan@epsilon.com -------------------------------------------------------------- ABSTRACTS AND BIOS: -------------------------------------------------------------- "SAS/Insight--Statistical Application Demo" by Steve Freitas, Chiron Diagnostics SAS/Insight is a Windows interactive exploratory statistics package offered as an add-on SAS product. This add-on requires no programming by the user, only a knowledge of some basic statistical techniques. This demonstration will focus on how to set up the Windows icon shortcut to Insight, exploring data in one/two dimensions, how to select observations based on certain preset criteria, how to hide or exclude outlying observations, and basic statistical analyses. Steve Freitas is a Senior Statistician at Chiron (soon to be Bayer) Diagnostics. Insight is used at Chiron as a tool for Scientists and Engineers to perform routine analyses on their data. By having this tool available, it frees up the statisticians to perform the more intricate statistical analyses required by the company. -------------------------------------------------------------- "Nonlinear Least Squares Analysis of HIV Viral Load Data" by Peter Gaccione, Independent Consultant This presentation will outline analyses which modeled HIV viral load data in a NIH sponsored clinical trial. An introduction to the biological model and its methods will be given, followed by the procedures used to fit the model and assess its viability. A further look will be given at relationships which might exist between baseline covariates and model parameter estimates. The emphasis of the talk will be on the use of SAS in doing the analyses. Many features of SAS/GRAPH and some straightforward invocations of the macro language were used during this work, and the presentation will focus on these as opposed to the results of the analyses. -------------------------------------------------------------- "An Introduction to Mixed Models in SAS" by William Simpson, Harvard Business School A mixed model is a type of general linear model where in addition to the usual (fixed) effects, at least one effect is considered random, i.e. the factor has levels that are considered a random sample from a larger universe of possible levels. Mixed models can be used to analyze problems where effects occur at different levels of analysis, such as classrooms and students, companies and employees, or experimental subjects and measurement times. In different applications, mixed models have been referred to by a number of different names: e.g. random coefficient regression, hierarchical linear models, multilevel models, and growth curve analysis. Some types of problems usually analyzed in a different way, such as repeated measures ANOVA, can also be formulated as mixed models. This talk is an introduction to mixed models with an emphasis on applications and on using PROC MIXED to estimate these models. No previous experience with mixed models or PROC MIXED is assumed, but a familiarity with regression and PROC GLM will be helpful. William Simpson is a Statistician/Programmer at the Faculty Research Computing Center of Harvard Business School, where he provides statistical consultation and programming for a wide variety of research projects. He has been a SAS programmer and consultant for 12 years. -------------------------------------------------------------- Directions to Epsilon, 50 Cambridge Street, Burlington, MA (you will see the Epsilon sign on the building on Route 128/I-95 around Exit 33A or 33B). 1) From Route 93, take exit 37B "Route 128/I-95 South". Follow Route 128/I-95 South for 4 miles, then take exit 33A "Route 3 South, Winchester." This brings you onto Cambridge Street. Proceed in the left lane and take a left at the second light. Epsilon is the first building on the left. 2) From Route 128/I-95 North, take exit 33B "Route 3 South, Winchester." This brings you onto Cambridge Street. Proceed in the left lane and take a left at the second light. Epsilon is the first building on the left. Parking: If the lot is filled, you can use the large shopping center located left side of the Epsilon facility. Public Transportation: The MBTA #350 bus leaves Alewife (red line) Station at 8:00AM and at 8:30AM for a 30-minute trip out to Cambridge Street. ----------------------------------------------------------- ----------------------------------------------------------- BASUG Web Site ----------------------------------------------------------- ----------------------------------------------------------- Our site on the World Wide Web always contains the latest information concerning upcoming meetings, classes, conferences, positions available, and consultants for hire. There is a new job opening added by local companies nearly every week. You may also retrieve Acrobat editions of "The BASUG Messenger", our new quarterly newsletter. Check us out at: http://world.std.com/~mmurphy/basug.html ----------------------------------------------------------- ----------------------------------------------------------- BASUG Contacts: ----------------------------------------------------------- ----------------------------------------------------------- The Boston Area SAS Users Group (BASUG) // e-mail: basug@world.std.com // WWW: http://world.std.com/~mmurphy/basug.html // US Mail: PO Box 253, Boston 02117 ----------------------------------------------------------- ----------------------------------------------------------- How to subscribe/unsubscribe from our message list ----------------------------------------------------------- ----------------------------------------------------------- The BASUG message list notifies subscribers of the latest plans concerning meetings, classes, conferences, and employment news of interest to local SAS users. To subscribe by e-mail: 1. Send a message to Majordomo@world.std.com 2. In the body of the message, put the following: subscribe basug To unsubscribe: 1. Send a message to Majordomo@world.std.com 2. In the body of the message, put the following: unsubscribe basug