A bootstrap simulation study in ARMA (p, q) structures

Abstract

In 1979 Efron proposed a new general statistical procedure known as 'Bootstrap', a computer-intensive method used when finite-sample theory is impossible or difficult to derive, or when only asymptotic theory is available. It is recommended in the estimation of measures of both location and scaIe for any statistical model without making any distributional assumptions about the data. This technique has been successfully used in various applied statistical problems, although not many applications have been reported in the area of time series. In this paper we present a new application of Bootstrap to time series. We consider a simulation study where artificial time series corresponding to AR(l), AR(2), MA(l), MA(2) and ARMA(1,l) structures were generated, covering important regions of the parameter space of each one of them. The conventional Box -Jenkins parametric estimators of the parameters are compared with the corresponding non-parametric Bootstrap estimators, obtained by 500 Bootstrap repetitions for each series.

Instantaneous Frequency Estimation Using Stochastic Calculus and Bootstrapping

“…Using the peak of the histogram, generated from the bootstrapping method for each estimated parameter, as the desired estimate of the unknown parameter, however, one is able to get good estimates at lower SNR. This was observed in this report and in some other applications (see Souza and Neto [25], Abutaleb [26]). This could be explained from the fact that the bootstrapping samples are almost the same as Monte Carlo simulations.…”

“…For a sampling interval ∆, the discrete WVD is given as (Boashash and Black [31]) (26) where M = (N − 1)/2, and N is the number of data points. Both formulae (25) and (26) are used in the literature.…”

Socially Responsible Investing and Sustainable Indices: A Sustainability Agenda

“…Also, augmented Dickey-Fuller test statistic was found to be −14.7242 with a p-value of .0000, and series is found to be stationary at level. The ARMA model is developed for Carbonex series by applying the criteria and steps specified by Tsay and Tiao (1984), Souza and Neto (1996) and Chan (1999). For model identification, the basic thumb rules were used such as exponential decay or damped, sine wave or both in ACF points to AR(p) model, significant spikes through first lag in ACF hints at MA(q) model and exponential decay means possible ARMA (p, q) model.…”

Electricity Load Forecasting With Artificial Neural Networks

“…Bootstrapping is a resampling method introduced by Efron [4]. Different methods have been proposed for bootstrapping in the context of time series [2], [7], [9], [14], [15]. In this article, the method based on residuals is used ([3], [9] and [5]).…”