Dataset Information

A novel imputation methodology for time series based on pattern sequence forecasting.

ABSTRACT: The Pattern Sequence Forecasting (PSF) algorithm is a previously described algorithm that identifies patterns in time series data and forecasts values using periodic characteristics of the observations. A new method for univariate time series is introduced that modifies the PSF algorithm to simultaneously forecast and backcast missing values for imputation. The imputePSF method extends PSF by characterizing repeating patterns of existing observations to provide a more precise estimate of missing values compared to more conventional methods, such as replacement with means or last observation carried forward. The imputation accuracy of imputePSF was evaluated by simulating varying amounts of missing observations with three univariate datasets. Comparisons of imputePSF with well-established methods using the same simulations demonstrated an overall reduction in error estimates. The imputePSF algorithm can produce more precise imputations on appropriate datasets, particularly those with periodic and repeating patterns.

SUBMITTER: Bokde N

PROVIDER: S-EPMC6223181 | biostudies-literature | 2018 Dec

REPOSITORIES: biostudies-literature

ACCESS DATA

Publications

A novel imputation methodology for time series based on pattern sequence forecasting.

Bokde Neeraj N Martínez Álvarez Francisco F Beck Marcus W MW Kulat Kishore K

Pattern recognition letters 20181201

The Pattern Sequence Forecasting (PSF) algorithm is a previously described algorithm that identifies patterns in time series data and forecasts values using periodic characteristics of the observations. A new method for univariate time series is introduced that modifies the PSF algorithm to simultaneously forecast and backcast missing values for imputation. The imputePSF method extends PSF by characterizing repeating patterns of existing observations to provide a more precise estimate of missing ...[more]

PMID: 30416234

Similar Datasets

Project description:In this paper, we introduce a novel model selection approach to time series forecasting. For linear stationary processes, such as AR processes, the direction of time is independent of the model parameters. By combining theoretical principles of time-reversibility in time series with conventional modeling approaches such as information criteria, we construct a criterion that employs the backwards prediction (backcast) as a proxy for the forecast. Hereby, we aim to adopt a theoretically grounded, data-driven approach to model selection. The novel criterion is named the backwards validated information criterion (BVIC). The BVIC identifies suitable models by trading off a measure of goodness-of-fit and a models ability to predict backwards. We test the performance of the BVIC by conducting experiments on synthetic and real data. In each experiment, the BVIC is examined in contrast to conventionally employed criteria. Our experimental results suggest that the BVIC has comparable performance as conventional information criteria. Specifically, in most of the experiments performed, we did not find statistically significant differences between the forecast error of the BVIC under certain parameterizations and that of the different information criteria. Nonetheless, it is worth emphasizing that the BVIC guarantees are established by design where the model order penalization term depends on strong mathematical properties of time-reversible time series forecasting properties and a finite data assessment. In particular, the penalization term is replaced by a weighted trade-off between functional dimensions pertaining to forecasting.That said, we observed that the BVIC recovered more accurately the real order of the underlying process than the other criteria, which rely on a static penalization of the model order. Lastly, leveraging the latter property we perform the assessment of the order model (or, memory) of time series pertaining to epileptic seizures recorded using electrocorticographic data. Our results provide converging evidence that the order of the model increases during the epileptic events.

Project description:Dynamic expression data, nowadays obtained using high-throughput RNA sequencing, are essential to monitor transient gene expression changes and to study the dynamics of their transcriptional activity in the cell or response to stimuli. Several methods for data selection, clustering and functional analysis are available; however, these steps are usually performed independently, without exploiting and integrating the information derived from each step of the analysis.Here we present FunPat, an R package for time series RNA sequencing data that integrates gene selection, clustering and functional annotation into a single framework. FunPat exploits functional annotations by performing for each functional term, e.g. a Gene Ontology term, an integrated selection-clustering analysis to select differentially expressed genes that share, besides annotation, a common dynamic expression profile.FunPat performance was assessed on both simulated and real data. With respect to a stand-alone selection step, the integration of the clustering step is able to improve the recall without altering the false discovery rate. FunPat also shows high precision and recall in detecting the correct temporal expression patterns; in particular, the recall is significantly higher than hierarchical, k-means and a model-based clustering approach specifically designed for RNA sequencing data. Moreover, when biological replicates are missing, FunPat is able to provide reproducible lists of significant genes. The application to real time series expression data shows the ability of FunPat to select differentially expressed genes with high reproducibility, indirectly confirming high precision and recall in gene selection. Moreover, the expression patterns obtained as output allow an easy interpretation of the results.A novel analysis pipeline was developed to search the main temporal patterns in classes of genes similarly annotated, improving the sensitivity of gene selection by integrating the statistical evidence of differential expression with the information on temporal profiles and the functional annotations. Significant genes are associated to both the most informative functional terms, avoiding redundancy of information, and the most representative temporal patterns, thus improving the readability of the results. FunPat package is provided in R/Bioconductor at link: http://sysbiobig.dei.unipd.it/?q=node/79.

Dataset Information

A novel imputation methodology for time series based on pattern sequence forecasting.

Publications

A novel imputation methodology for time series based on pattern sequence forecasting.

Similar Datasets

OmicsDI is part of the ELIXIR infrastructure

Tweets