We apply a recently proposed Bayesian model selection technique, known as stochastic model specification search, for characterising the nature of the trend in macroeconomic time series. We illustrate that the methodology can be quite successfully applied to discriminate between stochastic and deterministic trends. In particular, we formulate autoregressive models with stochastic trends componen...

Human syntactic processing shows many signs of taking place within a general-purpose shortterm memory. But this kind of memory is known to have a severely constrained storage capacity — possibly constrained to as few as three or four distinct elements. This article describes a model of syntactic processing that operates successfully within these severe constraints, by recognizing constituents i...

In recent years, research in nonlinear time series analysis has grown rapidly. Substantial empirical evidence of nonlinearities in economic time series fluctuations has been reported in the literature. Nonlinear time series models have the advantage of being able to capture asymmetries, jumps, and time irreversibility which are characteristics of many observed financial and economic time series...

In this paper, we consider using a semiparametric regression approach to modelling non-linear autoregressive time series. Based on a ®nite series approximation to non-parametric components, an adaptive selection procedure for the number of summands in the series approximation is proposed. Meanwhile, a large sample study is detailed and a small sample simulation for the Mackey±Glass system is pr...

We develop a dynamic Bayesian beta model for modeling and forecasting single time series of proportions. This work is related to the class of the so called dynamic generalized linear models (DGLM). We use non-conjugate priors and some forms of approximate Bayesian analysis, including Linear Bayesian estimation. Some applications to both real and simulated data are provided.

0 n −1 s t Suppose that there is some true model which generated our time series data, x , . . . , x . Thi rue model is not AR. But we do want to consider using AR models to describe our data, since they a f provide a flexible, estimable, and interpretable class of models. Although the AR models have only ew parameters, the true model presumably has a huge number of parameters (perhaps inf init...

A new framework for analyzing sequential or temporal data such as time series is proposed. It differs from other approaches by the special emphasis on the interpretability of the results, since interpretability is of vital importance for knowledge discovery, that is, the development of new knowledge (in the head of a human) from a list of discovered patterns. While traditional approaches try to...

Suppose that we are given a time series where consecutive samples are believed to come from a probabilistic source, that the source changes from time to time and that the total number of sources is fixed. Our objective is to estimate the distributions of the sources. A standard approach to this problem is to model the data as a hidden Markov model (HMM). However, since the data often lacks the ...

This paper presents the theoretical development of new threshold autoregressive models based on trended time series. The theoretical arguments underlying the models are outlined and a nonlinear economic model is used to derive the speci cation of the empirical econometric models. Estimation and testing issues are considered and analysed. Additionally we apply the models to the empirical investi...

Journal rankings of specific research fields are often used for evaluation purposes, both of authors and institutions. These rankings can be defined by means of several methods, as expert assessment, scholarly-based agreements, or by the ordering induced by a numeric index associated to the prestige of the journals. In order to be efficient and accepted by the research community, it must preser...