Parametric polyspectrum density estimation using the bootstrap method
Signal Processing
Hi-index | 0.00 |
The authors propose to apply bootstrap based techniques to investigate and improve the estimates of higher-order cumulants obtained from short data records. Algorithms for the calculation of the standard deviation and the confidence interval of cumulant estimates have been developed. Based on the algorithms, the authors describe a method for the estimation of risk function of various sampled cumulants, with the goal of choosing the estimator with best risk properties in the bootstrapping sense. Simulation results were obtained and are shown in tables.