Robust Functional Supervised Classification for Time Series

Alonso, Ana; Casado, David; López‐Pintado, Sara; Romo, Juan

doi:10.1007/s00357-014-9163-x

Cited by 9 publications

(6 citation statements)

References 25 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Explosive VAR processes have been shown to naturally arise in some disciplines, such as macroeconomics [56]. The matrices of coefficients in Scenario 5 vary with t, thus making the processes nonstationary despite verifying the stationarity conditions for each t. This scenario is a multivariate extension of the scenario of Simulation 2 in [57]. Timevarying VAR models have empirical applications in monetary policy [58] or in estimating fiscal and monetary interactions, among others [59].…”

Section: Experimental Designmentioning

confidence: 99%

F4: An All-Purpose Tool for Multivariate Time Series Classification

López-Oriona

Vilar

2021

Mathematics

View full text Add to dashboard Cite

We propose Fast Forest of Flexible Features (F4), a novel approach for classifying multivariate time series, which is aimed to discriminate between underlying generating processes. This goal has barely been addressed in the literature. F4 consists of two steps. First, a set of features based on the quantile cross-spectral density and the maximum overlap discrete wavelet transform are extracted from each series. Second, a random forest is fed with the extracted features. An extensive simulation study shows that F4 outperforms some powerful classifiers in a wide variety of situations, including stationary and nonstationary series. The proposed method is also capable of successfully discriminating between electrocardiogram (ECG) signals of healthy subjects and those with myocardial infarction condition. Additionally, despite lacking shape-based information, F4 attains state-of-the-art results in some datasets of the University of East Anglia (UEA) multivariate time series classification archive.

show abstract

Section: Experimental Designmentioning

confidence: 99%

F4: An All-Purpose Tool for Multivariate Time Series Classification

López-Oriona

Vilar

2021

Mathematics

View full text Add to dashboard Cite

show abstract

“…To generate this the series W i are constructed via an AR (5) model driven by i. i.d errors ∼ N(0, 1). The AR (5) coefficients are set to 0.5, 0.1, 0.1, 0.1, 0.1.…”

Section: Simulation: Matching Sub-regions In Univariate Time Seriesmentioning

confidence: 99%

“…Euclidean distance or Dynamic Time Warping distance [1,2] • Matching features of the series e.g. autocorrelation coefficients [3], Pearson correlation coefficients [4], periodograms [5], extreme value behavior [6] • Number of matching subsequences in the series [7] • Similarity of embedding distributions of the series [8] In this paper we consider distance measures for applications involving clustering, classification and related data mining tasks in time series, random fields and other forms of possibly non i. i.d data.…”

Section: Introductionmentioning

confidence: 99%

Kernel Distance Measures for Time Series, Random Fields and Other Structured Data

Das

Mhaskar

Cloninger

2021

Front. Appl. Math. Stat.

View full text Add to dashboard Cite

This paper introduces kdiff, a novel kernel-based measure for estimating distances between instances of time series, random fields and other forms of structured data. This measure is based on the idea of matching distributions that only overlap over a portion of their region of support. Our proposed measure is inspired by MPdist which has been previously proposed for such datasets and is constructed using Euclidean metrics, whereas kdiff is constructed using non-linear kernel distances. Also, kdiff accounts for both self and cross similarities across the instances and is defined using a lower quantile of the distance distribution. Comparing the cross similarity to self similarity allows for measures of similarity that are more robust to noise and partial occlusions of the relevant signals. Our proposed measure kdiff is a more general form of the well known kernel-based Maximum Mean Discrepancy distance estimated over the embeddings. Some theoretical results are provided for separability conditions using kdiff as a distance measure for clustering and classification problems where the embedding distributions can be modeled as two component mixtures. Applications are demonstrated for clustering of synthetic and real-life time series and image data, and the performance of kdiff is compared to competing distance measures for clustering.

show abstract

“…Within the multistate comparison group, the authors selected similarity-based states as states with the most similar baseline gonorrhea and chlamydia series using an integrated periodogram dissimilarity index (Colorado for gonorrhea and Oklahoma for chlamydia). 26,27 To adjust for possible regional variations, Rhode Island was selected-the only state in the northeast U.S. meeting the comparison criteria.…”

Section: Design and Samplementioning

confidence: 99%