PubMed Health⌕ Search
Home/Archive/PubMed

PubMed · 8901544

What is a linear process?

P J Bickel·P Bühlmann
View original source ↗Export metadata ↓

Abstract

We argue that given even an infinitely long data sequence, it is impossible (with any test statistic) to distinguish perfectly between linear and nonlinear processes (including slightly noisy chaotic processes). Our approach is to consider the set of moving-average (linear) processes and study its closure under a suitable metric. We give the precise characterization of this closure, which is unexpectedly large, containing nonergodic processes, which are Poisson sums of independent and identically distributed copies of a stationary process. Proofs of these results will appear elsewhere.

Explore related subjects

Linear ModelsNonlinear DynamicsJournal ArticleResearch Support, Non-U.S. Gov'tResearch Support, U.S. Gov't, Non-P.H.S.

Keep this discovery

Explore connections, maps & timelines

BibTeXRIS

P J Bickel, P Bühlmann. 1996-10-29. What is a linear process?. https://doi.org/10.1073/pnas.93.22.12128

Cite the original work for its findings. Save a collection to share your selection of sources.

KEEP EXPLORING

Related citations

PubMed2025

A robust transfer learning approach for high-dimensional linear regression to support integration of multi-source gene expression data.

Transfer learning aims to integrate useful information from multi-source datasets to improve the learning performance of target data. This can be effectively applied in genomics when we learn the gene associations in a target tissue, and data from other tissues can be integrated. However, heavy-tail distribution and outliers are common in genomics data, which poses challenges to the effectiveness of current transfer learning approaches. In this paper, we study the transfer learning problem under high-dimensional linear models with t-distributed error (Trans-PtLR), which aims to improve the estimation and prediction of target data by borrowing information from useful source data and offering robustness to accommodate complex data with heavy tails and outliers. In the oracle case with known transferable source datasets, a transfer learning algorithm based on penalized maximum likelihood and expectation-maximization algorithm is established. To avoid including non-informative sources, we propose to select the transferable sources based on cross-validation. Extensive simulation experiments as well as an application demonstrate that Trans-PtLR demonstrates robustness and better performance of estimation and prediction when heavy-tail and outliers exist compared to transfer learning for linear regression model with normal error distribution. Data integration, Variable selection, T distribution, Expectation maximization algorithm, Genotype-Tissue Expression, Cross validation.

Linear Models↗
PubMed1999

Computing the mechanical index.

A computational nonlinear beam propagation model was used to compute the water path and in situ fields of a phased array transducer operating at 2 MHz. The computational source was matched to the transducer's z = 10 cm focal plane field. Subsequent computed propagations considered this source operating at source amplitudes up to 1.49 MPa in a water medium and in a tissue medium with an attenuation of 0.3 dB cm(-1) MHz(-1). The mechanical index was calculated in three ways based on these computations: extrapolated from one low amplitude water path propagation, extrapolated from a series of water path propagations using the existing Output Display Standard protocol, and directly from a series of tissue path propagations. These computed results suggest that extrapolation from derated measurements of a low level water path field can provide mechanical index estimates which progressively overestimate the in situ values. At the highest source amplitude considered, the linearly extrapolated mechanical index was 29% higher than the mechanical index computed by the tissue path propagations. The Output Display Standard protocol offered improved accuracy but consistently underestimated the in situ values. The maximum error for the Output Display Standard protocol was 8%. A variation of the Output Display Standard protocol in which mechanical index estimates were obtained from the on-axis spatial peak in the derated temporal peak rarefactional curve was also considered. The maximum error for this method was 3%. The results considered here also demonstrated the feasibility of computational investigations of high intensity clinical propagations.

Linear Models↗
PubMed1998

A large class of models derived from generalized linear models.

Generalized linear models may be extended in several ways. This paper describes five such extensions: (i) generalized additive models; (ii) the use of quasi-likelihood; (iii) joint modelling of mean and dispersion; (iv) introduction of extra random components to give hierarchical generalized linear models; (v) modelling of correlated responses within subjects in longitudinal models. These extensions are largely independent, and so can be combined in many ways to produce a large class of models. Finally, a further extension to dynamic forms of the models is sketched.

Linear Models↗