J. M. Prada-Sánchez

Predicting using Box-Jenkins, nonparametric, and bootstrap techniques

In this article, a new semiparametric prediction system is presented for time series. The prediction method incorporated to the system consists of a nonparametric part that estimates the trend, a Box–Jenkins prediction of the residual series, and some bootstrap methodology to construct prediction intervals. Consistency of the estimators proposed for the autoregression function and the parameters in the Box–Jenkins model and the validity of a new bootstrap resampling plan adapted to autoregressive integrated models are proved. The Monte Carlo simulation study, as well as the applications to real data (carried out with the automated system, incorporating the method, developed for predicting concentration levels in the surroundings of a Spanish power station), show that this method outperforms other standard competitors.

Prediction of SO2 pollution incidents near a power station using partially linear models and an historical matrix of predictor‐response vectors

Atmospheric SO2 concentrations at sampling stations near the fossil fuel fired power station at As Pontes (La Coruna, Spain) were predicted using a model for the corresponding time series consisting of a self-explicative term and a linear combination of exogenous variables. In a supplementary simulation study, models of this kind behaved better than the corresponding pure self-explicative or pure linear regression models. Copyright

Multivariate statistical analysis of precipitation chemistry in northwestern Spain

Abstract149 samples of rainwater were collected in the proximity of a power station in northwestern Spain at three rainwater monitoring stations. We analyze the resulting data using multivariate statistical techniques. Firstly, the Principal Component Analysis shows that there are three main sources of pollution in the area (a marine source, a rural source and an acid source). The impact from pollution from these sources on the immediate environment of the stations is studied using Factorial Discriminant Analysis.

Time-series analysis for ambient concentrations

In this paper we present a dynamic system which has been implemented to predict, every 5 min, the ambient concentrations of SO2 in the neighbourhood of a power station run by ENDESA, the National Electricity Company of Spain, in As Pontes. This prediction task is very important in order to prevent a high ground-level of concentration of SO2. For forecasting we use a mixed model which has a parametric component and a nonparametric one. We also construct confidence intervals for future observations using bootstrap and classical techniques.

Bootstrapping the Chambers–Dunstan estimate of a finite population distribution function

Abstract The Chambers–Dunstan estimate of the distribution of a finite population is based on fitting a superpopulation model to the regression of the random variable of interest on a known auxiliary variable. In this paper, the asymptotic distribution of both the Chambers–Dunstan estimate and its bootstrap version are described. The bootstrap performed by resampling a smoothed recentred version of the empirical distribution of the fitting errors is proven to be consistent, and a simulation for which satisfactory results were obtained is described.

Bootstrapping the Dorfman–Hall–Chambers–Dunstan estimator of a finite population distribution function

In this paper, we describe a bootstrap method to estimate the bias, the variance and the distribution of the non-parametric Chambers–Dunstan estimator (named the Dorfman–Hall–Chambers–Dunstan (DHCD) estimator) prediction error. Bootstrapping is based on a bootstrap population constructed by sampling the empirical distribution of the superpopulation model recentred residuals. The prediction error of the DHCD estimator is shown to converge to a normal distribution and the consistency of the bootstrap procedure is shown by imitating formally results of the distribution of the prediction error relative to the bootstrap population. Also, the asymptotic form of the optimal bandwidth and a bootstrap estimator of such bandwidth are given for the DHCD estimator. This theory is illustrated by simulation results. †E-mail: wences@zmat.usc.es ‡E-mail: prada@zmat.usc.es

Forecasting Using a Semiparametric Model

In this paper we present a forecasting system that has been used to control the contamination in the surroundings of a Power Station in Northwestern Spain. The system provides forecastings of the immision levels every five minutes as well as confidence intervals for such levels.

Bootstrapping the mean of a symmetric population

Using Edgeworth expansions we compare the rates of convergence of the normal approximation and two bootstrap approaches for the sample mean of a symmetric population. In a simulation study we see how bad is the Monte Carlo approximation when bootstrapping the Edgeworth expansions.

Prediction with Additive Models—Simulation and Application with Real Data

Abstract In this paper we study the predictor behaviour of the additive model. The prediction equation is introduced as well as the computational considerations to select the smoothing parameters through cross-validation. The additive predictor is compared with a partially linear predictor in a broad simulation study and an application to a real case, prediction of the atmospheric concentration of SO2 in sample stations.