Statistical analysis of a parametric model for photometric signals

This communication studies a new model for photometric signals under high flux assumption. Photometric signals are modeled by Gaussian autoregressive processes having the same mean and variance denoted Constraint Gaussian AR Processes (CGARP´s). This model is first derived from the data asymptotic distribution under high flux assumption. The performance of the CGARP parameter estimators is then studied by comparing their mean square errors to the Cramer Rao lower bounds (CRLB´s). Asymptotic expressions are derived to approximate the CRLB´s for large values of the number of samples. Computer simulations confirm the validity of these expressions. The achievable performance for CGARP parameter estimation is compared to those obtained with the unconstraint model. The purpose of this model is to derive a Neyman Pearson detector for the change-point detection problem that arises in the extrasolar planets detection problem.