ABSTRACT
This chapter presents the fundamental concepts, models, and application procedures of conventional time series analysis. It enhances the conventional procedures through the incorporation of system identification theory, procedures for handling real data, and spectral analysis. The chapter illustrates the power of time series modeling to fisheries population assessment and management, with emphasis on input-output transfer functions as a generalized model of fishing mortality. The autoregressive model assumes that the current value of an output series can be described by weighted values of past observations and a random error component. The chapter discusses the BCDF model in light of other input-output models, and presents an alternate procedure for model identification of both univariate and bivariate systems. An important use of a time series model is to forecast future values of an output series. Missing values are typical in temporal fishery data. Estimation of missing values can be accomplished through a variety of procedures.
