Current status linear regression

We discuss estimators for the finite-dimensional regression parameter in the current status linear regression model. It is shown that, using a simple truncation device, one can construct $\sqrt{n}$-consistent and asymptotically normal estimates of the finite-dimensional regression parameter with an asymptotic covariance matrix that is arbitrarily close to the matrix of the information lower bound. We illustrate this with a simulation study and provide algorithms for computing the estimates and for selecting the bandwidth with a bootstrap method. The connection with results on the binary choice model in the econometric literature is also discussed.