Tags

Type your tag names separated by a space and hit enter

Use of empirical likelihood to calibrate auxiliary information in partly linear monotone regression models.
Stat Med 2014; 33(10):1713-22SM

Abstract

In statistical analysis, a regression model is needed if one is interested in finding the relationship between a response variable and covariates. When the response depends on the covariate, then it may also depend on the function of this covariate. If one has no knowledge of this functional form but expect for monotonic increasing or decreasing, then the isotonic regression model is preferable. Estimation of parameters for isotonic regression models is based on the pool-adjacent-violators algorithm (PAVA), where the monotonicity constraints are built in. With missing data, people often employ the augmented estimating method to improve estimation efficiency by incorporating auxiliary information through a working regression model. However, under the framework of the isotonic regression model, the PAVA does not work as the monotonicity constraints are violated. In this paper, we develop an empirical likelihood-based method for isotonic regression model to incorporate the auxiliary information. Because the monotonicity constraints still hold, the PAVA can be used for parameter estimation. Simulation studies demonstrate that the proposed method can yield more efficient estimates, and in some situations, the efficiency improvement is substantial. We apply this method to a dementia study.

Authors+Show Affiliations

Department of Biostatistics, University of Nebraska Medical Center, Omaha, NE, 68198, U.S.A.No affiliation info available

Pub Type(s)

Journal Article
Research Support, N.I.H., Extramural

Language

eng

PubMed ID

24323567

Citation

Chen, Baojiang, and Jing Qin. "Use of Empirical Likelihood to Calibrate Auxiliary Information in Partly Linear Monotone Regression Models." Statistics in Medicine, vol. 33, no. 10, 2014, pp. 1713-22.
Chen B, Qin J. Use of empirical likelihood to calibrate auxiliary information in partly linear monotone regression models. Stat Med. 2014;33(10):1713-22.
Chen, B., & Qin, J. (2014). Use of empirical likelihood to calibrate auxiliary information in partly linear monotone regression models. Statistics in Medicine, 33(10), pp. 1713-22. doi:10.1002/sim.6057.
Chen B, Qin J. Use of Empirical Likelihood to Calibrate Auxiliary Information in Partly Linear Monotone Regression Models. Stat Med. 2014 May 10;33(10):1713-22. PubMed PMID: 24323567.
* Article titles in AMA citation format should be in sentence-case
TY - JOUR T1 - Use of empirical likelihood to calibrate auxiliary information in partly linear monotone regression models. AU - Chen,Baojiang, AU - Qin,Jing, Y1 - 2013/12/09/ PY - 2013/02/08/received PY - 2013/11/13/revised PY - 2013/11/14/accepted PY - 2013/12/11/entrez PY - 2013/12/11/pubmed PY - 2014/12/15/medline KW - auxiliary information KW - empirical likelihood KW - isotonic regression KW - missing data SP - 1713 EP - 22 JF - Statistics in medicine JO - Stat Med VL - 33 IS - 10 N2 - In statistical analysis, a regression model is needed if one is interested in finding the relationship between a response variable and covariates. When the response depends on the covariate, then it may also depend on the function of this covariate. If one has no knowledge of this functional form but expect for monotonic increasing or decreasing, then the isotonic regression model is preferable. Estimation of parameters for isotonic regression models is based on the pool-adjacent-violators algorithm (PAVA), where the monotonicity constraints are built in. With missing data, people often employ the augmented estimating method to improve estimation efficiency by incorporating auxiliary information through a working regression model. However, under the framework of the isotonic regression model, the PAVA does not work as the monotonicity constraints are violated. In this paper, we develop an empirical likelihood-based method for isotonic regression model to incorporate the auxiliary information. Because the monotonicity constraints still hold, the PAVA can be used for parameter estimation. Simulation studies demonstrate that the proposed method can yield more efficient estimates, and in some situations, the efficiency improvement is substantial. We apply this method to a dementia study. SN - 1097-0258 UR - https://www.unboundmedicine.com/medline/citation/24323567/Use_of_empirical_likelihood_to_calibrate_auxiliary_information_in_partly_linear_monotone_regression_models_ L2 - https://www.ncbi.nlm.nih.gov/pmc/articles/pmid/24323567/ DB - PRIME DP - Unbound Medicine ER -