Date:21 June 2019, Friday
Location:E1-06-01, Faculty of Engineering, NUS
Time:03:00PM to 04:00PM
Measurement error arises ubiquitously from various fields including health sciences, epidemiological studies, survey research, economics, and so on. It has been a long standing concern in data analysis and has attracted extensive research interest over the past few decades. The effects of measurement error are complex and vary from problem to problem. While there are settings where measurement error effects are negligible, it has been well documented that ignoring measurement error in statistical analyses often yields erroneous or even misleading results. It is sensible to conduct a case-by-case examination in order to reach a valid statistical analysis for error-contaminated data. Although in practice both measurement error in covariates and misclassification in covariates may occur simultaneously, research attention in the literature has mainly focused on addressing either one of these problems separately but not both. In this talk, I will discuss issues pertinent to analysis of error-contaminated data and describe several methods of handling data with both measurement error and misclassification in covariates.