Identifying nonlinear covariate effects in semimartingale regression models (Q1262059): Difference between revisions

Let \(X_ t\) be a semimartingale which is either continuous or of counting process type and which satisfies the stochastic differential equation \[ dX_ t=Y_ t\alpha (t,Z_ t)dt+dM_ t, \] where Y and Z are predictable covariate processes, M is a martingale and \(\alpha\) is an unknown, nonrandom function. We study inference for \(\alpha\) by introducing an estimator for \[ {\mathcal A}(t,z)=\int^{z}_{0}\int^{t}_{0}\alpha (s,x)ds dx \] and deriving a functional central limit theorem for the estimator. The asymptotic distribution turns out to be given by a Gaussian random field that admits a representation as a stochastic integral with respect to a multiparameter Wiener process. This result is used to develop a test for independence of X from the covariate Z, a test for time-homogeneity of \(\alpha\), and a goodness-of- fit test for the proportional hazards model \(\alpha (t,z)=\alpha_ 1(t)\alpha_ 2(z)\) used in survival analysis.