Publications and Preprints

Estimation of parameters of two-dimensional sinusoidal signal in heavy-tailed errors
by
Swagata Nandi
In this paper, we consider a two-dimensional sinusoidal model observed in a additive random field. The proposed model has wide applications in statistical signal processing. The additive noise has mean zero but the variance may not be finite. We propose the least squares estimators to estimate the unknown parameters. It is observed that the least squares estimators are strongly consistent. We obtain the asymptotic distribution of the least squares estimators under the assumption that the additive errors are from a symmetric stable distribution. Some numerical experiments are performed to see how the results work for finite samples.

isid/ms/2012/05 [fulltext]

Click here to return to Preprints Page