A Diagnostic Procedure for Detecting Outliers in Linear State-Space Models

被引:12
作者
You, Dongjun [1 ]
Hunter, Michael [2 ]
Chen, Meng [1 ]
Chow, Sy-Miin [1 ]
机构
[1] Penn State Univ, State Coll, PA 16801 USA
[2] Georgia Inst Technol, Atlanta, GA 30332 USA
关键词
outlier detection; state space model; innovative outlier; additive outlier; PROCESSING SPEED; TIME; SHIFTS; IDENTIFICATION; SENSITIVITY; LIKELIHOOD; ATTENTION; CURVE;
D O I
10.1080/00273171.2019.1627659
中图分类号
O1 [数学];
学科分类号
0701 ; 070101 ;
摘要
Outliers can be more problematic in longitudinal data than in independent observations due to the correlated nature of such data. It is common practice to discard outliers as they are typically regarded as a nuisance or an aberration in the data. However, outliers can also convey meaningful information concerning potential model misspecification, and ways to modify and improve the model. Moreover, outliers that occur among the latent variables (innovative outliers) have distinct characteristics compared to those impacting the observed variables (additive outliers), and are best evaluated with different test statistics and detection procedures. We demonstrate and evaluate the performance of an outlier detection approach for multi-subject state-space models in a Monte Carlo simulation study, with corresponding adaptations to improve power and reduce false detection rates. Furthermore, we demonstrate the empirical utility of the proposed approach using data from an ecological momentary assessment study of emotion regulation together with an open-source software implementation of the procedures.
引用
收藏
页码:231 / 255
页数:25
相关论文
共 68 条
[1]  
[Anonymous], 1979, Optimal Filtering
[2]  
[Anonymous], 2008, WECHSLER ADULT INTEL
[3]  
[Anonymous], DYNR DYNAMIC MODELIN
[4]  
Aoki Masanao, 2013, State space modeling of time series
[5]   Influence diagnostics for linear longitudinal models [J].
Banerjee, M ;
Frees, EW .
JOURNAL OF THE AMERICAN STATISTICAL ASSOCIATION, 1997, 92 (439) :999-1005
[6]  
Bar-Shalom Y., 2002, Estimation with Applications to Tracking and Navigation, P453, DOI DOI 10.1002/0471221279
[7]   CHANGES IN PSYCHOLOGICAL STATE MEASURES AND TIME OF DAY [J].
BARTON, K ;
CATTELL, RB .
PSYCHOLOGICAL REPORTS, 1974, 35 (01) :219-222
[8]   INTERVENTION ANALYSIS WITH APPLICATIONS TO ECONOMIC AND ENVIRONMENTAL PROBLEMS [J].
BOX, GEP ;
TIAO, GC .
JOURNAL OF THE AMERICAN STATISTICAL ASSOCIATION, 1975, 70 (349) :70-79
[9]  
Brookner E., 1998, Tracking and Kalman filtering made easy, DOI DOI 10.1002/0471224197
[10]  
Browne MW, 2005, MULTIVAR APPL SER, P415