The Box-Jenkins approach to time series analysis
The Box-Jenkins approach to time series analysis and forecasting : principles and applications
The cross-validation method in the polynomial regression.
The Effect of Parameter Estimation Errors on the Variance of Forecasts
The effects of series on cointegration inference.
The efficiency of estimates in stationary autoregressive series
The empirical TES methodology: Modeling empirical time series.
The Exact Likelihood Function for a Space Time Model.
The failure of orthogonality under nonstationarity: should we care about it?
The finite automata approaches in stringology
We present an overview of four approaches of the finite automata use in stringology: deterministic finite automaton, deterministic simulation of nondeterministic finite automaton, finite automaton as a model of computation, and compositions of finite automata solutions. We also show how the finite automata can process strings build over more complex alphabet than just single symbols (degenerate symbols, strings, variables).
The Invertibility of Sampled and Aggregated ARM A Models.
The Kolmogorov-Smirnov distance as criterion of choice of estimators with application to the first order auto-regressive case : a Monte Carlo study
The relationship between stock markets of major developed countries and Asian emerging markets.
The Serial Correlation Structure for a Random Process with Steps.
The smoothness criterion as a trend diagnostic.
The truncated least squares method for a class of autoregressive models.
Threshold moving average model
Time series analysis: recursive methods and their modifications for time series with outliers and missing observations.
The recursive methods are popular in time series analysis since they are computationally efficient and flexible enough to treat various changes in character of data. This paper gives a survey of the most important type of these methods including their classification and relationships existing among them. Special attention is devoted to i) robustification of some recursive methods, capable of facing outliers in time series, and ii) modifications of recursive methods for time series with missing observations....
Time series model identification by estimating information, memory and quantiles.
This paper applies techniques of Quantile Data Analysis to non-parametrically analyze time series functions such as the sample spectral density, sample correlations and sample partial correlations. The aim is to identify the memory type of an observed time series, and thus to identify parametric time domain models that fit an observed time series. Time series models are usually tested for adequacy by testing if their residuals are white noise. It is proposed that an additional criterion of fit for...
Time series modelling of the Kobe-Osaka earthquake recordings.