likelihood criterion
1.Calculating the translation of beacon based on discrete Fourier transform(DFT) and maximum likelihood criterion,need solve two nonlinear equations,and the linearly approximated result shows that the total translation error is below 0.5 pixel,the translation error in X direction is 3.3%,and the translation error in Y the direction is 2.7%,can meet the need of deep space optical communication,so it is a feasible scheme.
2.Anderson , Amemiva and Fujisawa et al scholars extended the growth curve model into the one with random effects and considered the likelihood ratio criterion (LRC) for its mean structure ,where the random vectors follow normal distribution on the assumptions that random effects and random errors are mutually independent[1][2] .
4.To get high numerical stability and computational efficiency,a new AIC criterion and recursive parameter estimation methods are developed which make both the maximum likelihood method and the optimal criterion for model choice of SISO system very simple. For MIMO system,a new AIC criterion is given and the computation formulas of the cost function is simplified. The results of aerodynamic model identification show that the new method can give accurate model structures of unknown system.
5.In the paper correlation identification, correlation filtering, criterion of AIC and maximum likelihood estimation are combined to detect the fault of a DC motor system.
6.The problem of model fitting is one of the most important ones in statistical inference. Regression models in which random influence is normal distribution can be discriminated easily by AIC criterion. This paper discusses the entropy maximization principle, the least AIC criterion and the maximum likelihood function, lists some representative regression models, points out some problems which we must pay attention to in the procedure of model fitting.
