An extended Kalman filter algorithm for aerodynamic parameter identification from missile postflight data is developed and verified for realistic test conditions. The algorithm includes a general purpose six-degrees- of-freedom missile airframe model suitable for representing a variety of missile configurations. Verification studies consider low order linear aerodynamic models and higher order models with extensive nonlinear and pitch-yaw coupling effects. The sensitivity of filter performance to initial conditions, measurement data rate and accuracy, input selection, and modeling errors is investigated. A structure identification technique is used to select the most probable aerodynamic model for a given data set from a group of candidate models. In addition, actual flight test data from a complex aerodynamically controlled vehicle is processed with the filter algorithm. The resulting identified model is shown to be an improvement over the preflight model.