Prediction for time series in the fraction-of-time probability framework

摘要: Abstract The aim of this work is to introduce the concept quantile and propose its prediction algorithms using fraction-of-time probability approach. In such an approach, unlike classical one based on stochastic processes, statistical functions concepts are defined starting from a single observed time series instead ensemble realizations process. Two without any distributional assumption proposed. former devoted deal with statistics not depending (stationary case) whereas latter considers that depend (nonstationary case). Convergence estimation accuracy issues considered in paper checking usual mixing assumptions, used Moreover, applications design constant false-alarm rate radar processors analysis real financial data presented.