An Empirical Study of a Rank-Based Estimate for Cell Lineage Data

• Activity Number: 525
• Type: Contributed
• Date/Time: Wednesday, August 3, 2011 : 10:30 AM to 12:20 PM
• Sponsor: Section on Nonparametric Statistics
• Author(s): Tamer Elbayoumi*+ and Jeffrey Terpstra
• Companies: Western Michigan University and Western Michigan University
• Keywords: Rank-Based Estimates ; Least Squares Estimate ; Bifurcating Autoregressive ; Cell Lineage Data ; Sample Relative Efficiency ; Aberrant Observations

Abstract

The presence of aberrant observations (i.e. outliers) in cell lineage data is quite common. As such, it is desirable to have an outlier-resistant estimation procedure as an alternative to least squares estimation (maximum likelihood estimation under normality). In this work, we consider rank-based estimates of the parameters of a first order bifurcating autoregressive [BAR(1)] model. The BAR(1) model was proposed by Cowan and Staudte (1986) for cell lineage data. In it, each line of descendents follows a first order autoregressive [AR(1)] model and allows sister cells from the same mother to be correlated. Real examples and a simulation study are performed in order to examine the behavior of these rank-based estimation procedures. More specifically, we compute finite sample relative efficiencies with respect to least squares estimate. The results indicate that the rank-based estimation procedures are more efficient when outlying observations are present.