Application of the Multivariate Prediction Method to Time Series

* This work was financially supported by RFBR-04-01-00858.

Nets: Network estimation for time series

This work proposes novel network analysis techniques for multivariate time series.We define the network of a multivariate time series as a graph where verticesdenote the components of the process and edges denote non zero long run partialcorrelations. We then introduce a two step LASSO procedure, called NETS, toestimate high dimensional sparse Long Run Partial Correlation networks. This approachis based on a VAR approximation of the process and allows to decomposethe long run linkages into the contribution of the dynamic and contemporaneousdependence relations of the system. The large sample properties of the estimatorare analysed and we establish conditions for consistent selection and estimation ofthe non zero long run partial correlations. The methodology is illustrated with anapplication to a panel of U.S. bluechips.

Why Italy and not Spain?: comparing two industrialization processes from a dissagregate time-series perspective

This study presents new evidence concerning the uneven processes of industrialization innineteenth century Spain and Italy based on a disaggregate analysis of the productivesectors from which the behaviour of the aggregate indices is comprised. The use of multivariate time-series analysis techniques can aid our understanding and characterization of these two processes of industrialization. The identification of those sectors with key rolesin leading industrial growth provides new evidence concerning the factors that governed thebehaviour of the aggregates in the two economies. In addition, the analysis of the existenceof interindustry linkages reveals the scale of the industrialization process, and wheresignificant differences exist, accounts for many of the divergences recorded in the historiography for the period 1850-1913.

Why Italy and not Spain?: comparing two industrialization processes from a dissagregate time-series perspective

This study presents new evidence concerning the uneven processes of industrialization innineteenth century Spain and Italy based on a disaggregate analysis of the productivesectors from which the behaviour of the aggregate indices is comprised. The use of multivariate time-series analysis techniques can aid our understanding and characterization of these two processes of industrialization. The identification of those sectors with key rolesin leading industrial growth provides new evidence concerning the factors that governed thebehaviour of the aggregates in the two economies. In addition, the analysis of the existenceof interindustry linkages reveals the scale of the industrialization process, and wheresignificant differences exist, accounts for many of the divergences recorded in the historiography for the period 1850-1913.

ESTIMATING TEMPORAL ASSOCIATIONS IN ELECTROCORTICOGRAPHIC (ECoG) TIME SERIES WITH FIRST ORDER PRUNING

Granger causality (GC) is a statistical technique used to estimate temporal associations in multivariate time series. Many applications and extensions of GC have been proposed since its formulation by Granger in 1969. Here we control for potentially mediating or confounding associations between time series in the context of event-related electrocorticographic (ECoG) time series. A pruning approach to remove spurious connections and simultaneously reduce the required number of estimations to fit the effective connectivity graph is proposed. Additionally, we consider the potential of adjusted GC applied to independent components as a method to explore temporal relationships between underlying source signals. Both approaches overcome limitations encountered when estimating many parameters in multivariate time-series data, an increasingly common predicament in today's brain mapping studies.

Reference list of 120 datasets from time series station Payerne used for exploratory search

The analysis of time-dependent data is an important problem in many application domains, and interactive visualization of time-series data can help in understanding patterns in large time series data. Many effective approaches already exist for visual analysis of univariate time series supporting tasks such as assessment of data quality, detection of outliers, or identification of periodically or frequently occurring patterns. However, much fewer approaches exist which support multivariate time series. The existence of multiple values per time stamp makes the analysis task per se harder, and existing visualization techniques often do not scale well. We introduce an approach for visual analysis of large multivariate time-dependent data, based on the idea of projecting multivariate measurements to a 2D display, visualizing the time dimension by trajectories. We use visual data aggregation metaphors based on grouping of similar data elements to scale with multivariate time series. Aggregation procedures can either be based on statistical properties of the data or on data clustering routines. Appropriately defined user controls allow to navigate and explore the data and interactively steer the parameters of the data aggregation to enhance data analysis. We present an implementation of our approach and apply it on a comprehensive data set from the field of earth bservation, demonstrating the applicability and usefulness of our approach.

Hidden state models for time series

Amongst all the objectives in the study of time series, uncovering the dynamic law of its generation is probably the most important. When the underlying dynamics are not available, time series modelling consists of developing a model which best explains a sequence of observations. In this thesis, we consider hidden space models for analysing and describing time series. We first provide an introduction to the principal concepts of hidden state models and draw an analogy between hidden Markov models and state space models. Central ideas such as hidden state inference or parameter estimation are reviewed in detail. A key part of multivariate time series analysis is identifying the delay between different variables. We present a novel approach for time delay estimating in a non-stationary environment. The technique makes use of hidden Markov models and we demonstrate its application for estimating a crucial parameter in the oil industry. We then focus on hybrid models that we call dynamical local models. These models combine and generalise hidden Markov models and state space models. Probabilistic inference is unfortunately computationally intractable and we show how to make use of variational techniques for approximating the posterior distribution over the hidden state variables. Experimental simulations on synthetic and real-world data demonstrate the application of dynamical local models for segmenting a time series into regimes and providing predictive distributions.

An investigation of imputation methods for discrete databases and multi-variate time series

When it comes to information sets in real life, often pieces of the whole set may not be available. This problem can find its origin in various reasons, describing therefore different patterns. In the literature, this problem is known as Missing Data. This issue can be fixed in various ways, from not taking into consideration incomplete observations, to guessing what those values originally were, or just ignoring the fact that some values are missing. The methods used to estimate missing data are called Imputation Methods. The work presented in this thesis has two main goals. The first one is to determine whether any kind of interactions exists between Missing Data, Imputation Methods and Supervised Classification algorithms, when they are applied together. For this first problem we consider a scenario in which the databases used are discrete, understanding discrete as that it is assumed that there is no relation between observations. These datasets underwent processes involving different combina- tions of the three components mentioned. The outcome showed that the missing data pattern strongly influences the outcome produced by a classifier. Also, in some of the cases, the complex imputation techniques investigated in the thesis were able to obtain better results than simple ones. The second goal of this work is to propose a new imputation strategy, but this time we constrain the specifications of the previous problem to a special kind of datasets, the multivariate Time Series. We designed new imputation techniques for this particular domain, and combined them with some of the contrasted strategies tested in the pre- vious chapter of this thesis. The time series also were subjected to processes involving missing data and imputation to finally propose an overall better imputation method. In the final chapter of this work, a real-world example is presented, describing a wa- ter quality prediction problem. The databases that characterized this problem had their own original latent values, which provides a real-world benchmark to test the algorithms developed in this thesis.

A multivariate approach for processing magnetization effects in triggered event-related functional magnetic resonance imaging time series

Triggered event-related functional magnetic resonance imaging requires sparse intervals of temporally resolved functional data acquisitions, whose initiation corresponds to the occurrence of an event, typically an epileptic spike in the electroencephalographic trace. However, conventional fMRI time series are greatly affected by non-steady-state magnetization effects, which obscure initial blood oxygen level-dependent (BOLD) signals. Here, conventional echo-planar imaging and a post-processing solution based on principal component analysis were employed to remove the dominant eigenimages of the time series, to filter out the global signal changes induced by magnetization decay and to recover BOLD signals starting with the first functional volume. This approach was compared with a physical solution using radiofrequency preparation, which nullifies magnetization effects. As an application of the method, the detectability of the initial transient BOLD response in the auditory cortex, which is elicited by the onset of acoustic scanner noise, was used to demonstrate that post-processing-based removal of magnetization effects allows to detect brain activity patterns identical with those obtained using the radiofrequency preparation. Using the auditory responses as an ideal experimental model of triggered brain activity, our results suggest that reducing the initial magnetization effects by removing a few principal components from fMRI data may be potentially useful in the analysis of triggered event-related echo-planar time series. The implications of this study are discussed with special caution to remaining technical limitations and the additional neurophysiological issues of the triggered acquisition.

Multivariate analyses of variance and covariance for simulation studies involving normal time series

Photocopy.

An application of two techniques for the analysis of short, multivariate non-stationary time-series of Mauritanian trawl survey data

Min/max autocorrelation factor analysis (MAFA) and dynamic factor analysis (DFA) are complementary techniques for analysing short (> 15-25 y), non-stationary, multivariate data sets. We illustrate the two techniques using catch rate (cpue) time-series (1982-2001) for 17 species caught during trawl surveys off Mauritania, with the NAO index, an upwelling index, sea surface temperature, and an index of fishing effort as explanatory variables. Both techniques gave coherent results, the most important common trend being a decrease in cpue during the latter half of the time-series, and the next important being an increase during the first half. A DFA model with SST and UPW as explanatory variables and two common trends gave good fits to most of the cpue time-series. (c) 2004 International Council for the Exploration of the Sea. Published by Elsevier Ltd. All rights reserved.