In this article we consider a real process xTwhich is composed of a real function mT() and a white brownian noise UT. where the real function mT() represents a linear combination of known periodic functions ₁,….., k and some unknown parameters =(₁,...,k). We use the maximum likelihood method to estimate the unknown parameters. We proved a theorem which gives an upper bound for the probability that the absolute value of the Euclidian distance between the parameters and its estimators exceeds a small positive value ɛ. Some examples of periodic functions and the convergence of its sequence of means are given. Computation is performed for periodic functions i's of the form i(x)=sin ix. As application of periodic function we exposed MACHO project in the domain of astronomy.
View on SCiNiTO
Top