The primary goal of this continuing grant is to develop new statistical methods that handle missing data to better understand the mechanisms of neurological disorders, particularly stroke. The statistical problems to be investigated in this proposal arise from NIH-funded Northern Manhattan Stroke Study (NOMASS) in which the Principal Investigator has been participating. NOMASS started as a cross-sectional study for first time stroke patients, and was then extended to a case-control study by recruiting community controls via Random digit dialing method. Finally, both cases and controls are followed to observe recurrent stroke for cases and initial stroke for controls. We will consider missing data problems, which arise in each stage of design: cross-sectional, case-control, and prospective. Although these problems have frequently arisen in other fields as well, satisfactory solutions have not, as of yet, been found. ? ?
| Lee, Hye-Seung; Cho Paik, Myunghee; Lee, Joseph H (2009) Estimating a multivariate familial correlation using joint models for canonical correlations: application to memory score analysis from familial Hispanic Alzheimer's disease study. Biometrics 65:463-9 |
| Paik, Myunghee Cho; Wang, Cuiling (2009) HANDLING MISSING DATA BY DELETING COMPLETELY OBSERVED RECORDS. J Stat Plan Inference 139:2341-2350 |
| Lee, Hye-Seung; Paik, Myunghee Cho; Lee, Joseph H (2008) Genotype-adjusted familial correlation analysis using three generalized estimating equations. Stat Med 27:5471-83 |
| Cho Paik, Myunghee (2004) Nonignorable missingness in matched case-control data analyses. Biometrics 60:306-14 |
| Lin, I F; Paik, M C (2001) Matched case-control data analysis with selection bias. Biometrics 57:1106-12 |
| Paik, M C; Sacco, R; Lin, I F (2000) Bivariate binary data analysis with nonignorably missing outcomes. Biometrics 56:1145-56 |
