Colloquium

Colloquium


Statistics Colloquium
November 9, 11 a.m.
Alavi Commons Room, 6625 Everett Tower

A Framework for Statistical Methods Research

Mireya Diaz
Division Chief, Division of Epidemiology and Biostatistics
WMU Homer Stryker M.D. School of Medicine

Abstract:
Medical research offers a landscape apt for the development of statistical methods as well as their systematic performance evaluation. This is due to the richness of complexity surrounding medical information that sometimes is dealt with in a simplified fashion. Taking into consideration all the features present in a given dataset most likely would require extensions to existing methods or development of new approaches. In other instances, the utilization of particular methods is widely spread before their performance is thoroughly assessed making it difficult to refrain the use of suboptimal techniques. In this presentation I will walk you through these settings. For example, we will examine the performance of the bivariate random effects model for meta-analyses of diagnostic accuracy in a wide range of scenarios via staggered factorial designs, and illustrate how the results from these simulations challenge the customary practice of its indiscriminate use particularly when few studies are available.

Bio:
Mireya Diaz, PhD, has over 15 years of statistical consulting experience. She received her doctorate degree in Biostatistics from Case Western Reserve University. There she served as an Assistant Professor of Biostatistics, member of two NIH-funded biostatistics core - one as Director - and Director of the Biostatistics program. While in this position she also worked as a methodologist for the Guidelines Department of the American Urological Association providing the methodological rigor for the medical guidelines dealing with management of VUR in children. Then she accepted a position as Associate Scientist and Director of Biostatistics of the Vattikuti Urology Institute where she worked with the Menon's pioneering team of Robotic Prostatectomy generating the evidence for the comparative effectiveness of the robotic approach. Her research endeavors and contributions to science focus in the development, application, and evaluation of statistical methods for effectiveness research with large observational databases including disease registries, administrative claims, and EMRs; meta-analyses of complex outcome structures; health technology assessment and medical guidelines development; and correlated data.

Education:
   PhD 2002, Biostatistics, Case Western Reserve University
   MSc 1999, Biomedical Engineering, Case Western Reserve University
   BSc 1994, Electrical Engineering, Universidad Metropolitana

---
All statistics graduate students are expected to attend.

.

Colloquium schedule

Past colloquiums

 

Department of Statistics
3304 Everett Tower
Western Michigan University
Kalamazoo MI 49008-5152 USA
(269) 387-1420 | (269) 387-1419 Fax
stat-webmaster@wmich.edu