WEIGHTED ESTIMATING EQUATIONS WITH NONIGNORABLY MISSING RESPONSE DATA

We propose weighted estimating equations for data with nonignorable no nresponse in order to reduce the bias that can occur with a complete c ase analysis. A survey concerning medical practice guidelines, malprac tice litigation, and settlement provides the framework. The survey was sent to recipients in two waves: those who responded on the first or second wave are used to estimate a nonignorable nonresponse model, whi le the fraction of recipients who never responded is used to allow the percentage of missing data to change with each wave. We use the struc ture of the GEE of Liang and Zeger (1986, Biometrika 73, 13-22), addin g weights equal to the inverse probability of being observed. We prese nt simulations demonstrating the bias that can occur with an unweighte d analysis and use the survey data to illustrate the methods.