An MDL approach to the climate segmentation problem

Lu, QiQi; Lund, Robert; Lee, Thomas C.

doi:10.1214/09-aoas289

Cited by 67 publications

(56 citation statements)

References 28 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Given the successes of Davis et al (2006Davis et al ( , 2008, Lee (2001) and Lu et al (2010), we develop a genetic algorithm (GA) to minimize (7) efficiently. However, we note that our GA has some critical differences from those developed in these references.…”

Section: Minimization Of (7) Using Genetic Algorithmsmentioning

confidence: 99%

“…This constrain has also been used by Davis et al (2006Davis et al ( , 2008 and Lu et al (2010). However, we note that the chromosome representation adopted here is fundamentally different than those in Davis et al (2006Davis et al ( , 2008 and Lu et al (2010): in their work the length of a chromosome is always fixed to be n, while ours is of variable length. Consequently, the method for initializing the first generation, and the definitions for crossover and mutation are necessarily different in our implementation.…”

Section: Accepted Manuscriptmentioning

confidence: 99%

See 1 more Smart Citation

Estimation of Multiple-Regime Threshold Autoregressive Models With Structural Breaks

Yau

Tang

Lee

2015

Journal of the American Statistical Association

Self Cite

View full text Add to dashboard Cite

The threshold autoregressive (TAR) model is a class of nonlinear time series models that have been widely used in many areas. Due to its nonlinear nature, one major difficulty in fitting a TAR model is the estimation of the thresholds. As a first contribution, this paper develops an automatic procedure to estimate the number and values of the thresholds, as well as the corresponding AR order and parameter values in each regime. These parameter estimates are defined as the minimizers of an objective function derived from the minimum description length (MDL) principle. A genetic algorithm is constructed to efficiently solve the associated minimization problem. The second contribution of this paper is the extension of this framework to piecewise TAR modeling; i.e., the time series is partitioned into different segments for which each segment can be adequately modeled by a TAR model, while models from adjacent segments are different. For such piecewise TAR modeling, a procedure is developed to estimate the number and locations of the breakpoints, together with all other parameters in each segment. Desirable theoretical results are derived to lend support to the proposed methodology. Simulation experiments and an application to an U.S. GNP data are used to illustrate the empirical performances of the methodology.

show abstract

Section: Minimization Of (7) Using Genetic Algorithmsmentioning

confidence: 99%

Section: Accepted Manuscriptmentioning

confidence: 99%

Estimation of Multiple-Regime Threshold Autoregressive Models With Structural Breaks

Yau

Tang

Lee

2015

Journal of the American Statistical Association

Self Cite

View full text Add to dashboard Cite

show abstract

“…From MDL standpoint, any probability distribution is considered for its ability to describe the data and does not have to be identical to the distribution underlying the data-generating mechanism. The connection between MDL and statistical analysis has matured with the work of Barron et al (1998), Hansen andYu (2001), Lee (2000Lee ( , 2001; Lu et al (2010) and Hansen and Yu (2003). Hansen andYu (2001, 2003) presented various frameworks in which the MDL principle can be applied to (generalized) linear models with independent data.…”

Section: Among Others)mentioning

confidence: 99%

Minimum description length principle for linear mixed effects models

Li¹,

Yao

Craiu³

et al. 2014

STAT SINICA

View full text Add to dashboard Cite

AbstractsThe minimum description length (MDL) principle originated from data compression literature and has been considered for deriving statistical model selection procedures.Most of the existing methods that use the MDL principle focus on models with independent data, particularly in the context of linear regression. This paper considers data with repeated measurements and studies the selection of fixed effect covariates for linear mixed effect models. We propose a class of MDL procedures which incorporate the dependence structure within individual or cluster and uses data-adaptive penalties that suit both finite and infinite dimensional data generating mechanisms. Theoretical justifications are provided from both data compression and statistical perspectives, where the covariance of random effects is treated as known or estimated by maximum likelihood. Numerical experiments are conducted to demonstrate the usefulness of the proposed MDL procedure and the influence of the estimated covariance, and an application to U.S. EPA data for air quality control is provided.

show abstract

“…Recent work on detecting mean shift in climatic series includes [1,18,19,28]. [28] also provided a good literature review on the matter, while [25] provided a review and comparison of less recent techniques.…”

Section: Introductionmentioning

confidence: 99%

Detecting change-points in extremes

Dupuis

Sun

Wang

2015

Statistics and Its Interface

View full text Add to dashboard Cite

Even though most work on change-point estimation focuses on changes in the mean, changes in the variance or in the tail distribution can lead to more extreme events. In this paper, we develop a new method of detecting and estimating the change-points in the tail of multiple time series data. In addition, we adapt existing tail change-point detection methods to our specific problem and conduct a thorough comparison of different methods in terms of performance on the estimation of change-points and computational time. We also examine three locations on the U.S. northeast coast and demonstrate that the methods are useful for identifying changes in seasonally extreme warm temperatures.

show abstract

An MDL approach to the climate segmentation problem

Cited by 67 publications

References 28 publications

Estimation of Multiple-Regime Threshold Autoregressive Models With Structural Breaks

Estimation of Multiple-Regime Threshold Autoregressive Models With Structural Breaks

Minimum description length principle for linear mixed effects models

Detecting change-points in extremes

Contact Info

Product

Resources

About