Pena’s statistic for the Liu regression. (English) Zbl 07192669
Summary: In fitting regression model, one or more observations may have substantial effects on estimators. These unusual observations are precisely detected by a new diagnostic measure, Pena’s statistic. In this article, we introduce a type of Pena’s statistic for each point in Liu regression. Using the forecast change property, we simplify the Pena’s statistic in a numerical sense. It is found that the simplified Pena’s statistic behaves quite well as far as detection of influential observations is concerned. We express Pena’s statistic in terms of the Liu leverages and residuals. The normality of this statistic is also discussed and it is demonstrated that it can identify a subset of high Liu leverage outliers. For numerical evaluation, simulated studies are given and a real data set has been analysed for illustration.
MSC:
| 62J20 | Diagnostics, and linear inference and regression |
| 62J12 | Generalized linear models (logistic models) |
References:
| [1] | Belsley DA, Kuh E, Welsch RE. Regression diagnostics, identifying influential data and sources of collinearity. New York: Wiley; 1980. [Crossref], [Google Scholar] · Zbl 0479.62056 |
| [2] | Cook RD, Weisberg S. Residuals and influence in regression. New York: Chapman & Hall; 1982. [Google Scholar] · Zbl 0564.62054 |
| [3] | Atkinson AC. Plots, transformations and regression. Oxford: Clean-don Press; 1985. [Google Scholar] · Zbl 0582.62065 |
| [4] | Chatterjee S, Hadi AS. Sensitivity analysis in linear regression. New York: John Wiley & Sons; 1988. [Crossref], [Google Scholar] · Zbl 0648.62066 |
| [5] | Ullah MA, Pasha GR.The origin and developments of influence measures in regression. Pak J Statist. 2009;25:295-307. [Web of Science ®], [Google Scholar] · Zbl 1509.62278 |
| [6] | Pena D.A new statistic for influence in linear regression. Technometrics. 2005;47:1-12. doi: 10.1198/004017004000000662[Taylor & Francis Online], [Web of Science ®], [Google Scholar] |
| [7] | Jahufer A, Jianbao C.Assessing global influential observations in modified ridge regression. Statist Probab Lett. 2009;79:513-518. doi: 10.1016/j.spl.2008.09.019[Crossref], [Web of Science ®], [Google Scholar] · Zbl 1155.62051 |
| [8] | Amanullah M, Pasha GR, Aslam M.Local influence diagnostics in the modified ridge regression. Commun Statist Theory Methods. 2013;42:1851-1869. doi: 10.1080/03610926.2011.597920[Taylor & Francis Online], [Web of Science ®], [Google Scholar] · Zbl 1268.62083 |
| [9] | Swindel BF.Good ridge estimators based on prior information. Commun Statist Theory Methods. 1976;5:1065-1075. doi: 10.1080/03610927608827423[Taylor & Francis Online], [Web of Science ®], [Google Scholar] · Zbl 0342.62035 |
| [10] | Walker E, Birch JB.Influence measures in ridge regression. Technometrics. 1988;30:221-227. doi: 10.1080/00401706.1988.10488370[Taylor & Francis Online], [Web of Science ®], [Google Scholar] |
| [11] | Shi L, Wang X.Local influence in ridge regression. Comput Statist Data Anal. 1999;31:341-353. doi: 10.1016/S0167-9473(99)00019-5[Crossref], [Web of Science ®], [Google Scholar] · Zbl 0990.62055 |
| [12] | Hoerl AE, Kennard RW.Ridge regression: biased estimation for non-orthogonal problems. Technometrics. 1970;12:55-67. doi: 10.1080/00401706.1970.10488634[Taylor & Francis Online], [Web of Science ®], [Google Scholar] · Zbl 0202.17205 |
| [13] | Hoerl AE, Kennard RW.Ridge regression: applications to nonorthogonal problems. Technometrics. 1970;12:69-82. doi: 10.1080/00401706.1970.10488635[Taylor & Francis Online], [Web of Science ®], [Google Scholar] · Zbl 0202.17206 |
| [14] | Kejian L.A new class of blased estimate in linear regression. Commun Statist Theory Methods. 1993;22:393-402. doi: 10.1080/03610929308831027[Taylor & Francis Online], [Web of Science ®], [Google Scholar] · Zbl 0784.62065 |
| [15] | Amanullah M, Pasha GR, Aslam M.Assessing influence on the Liu estimates in linear regression models. Commun Statist Theory Methods. 2013;42:2400-2416. [Web of Science ®], [Google Scholar] |
| [16] | Emami H, Emami M.New influence diagnostics in ridge regression. J Appl Statist. 2016;43:476-489. doi: 10.1080/02664763.2015.1070804[Taylor & Francis Online], [Web of Science ®], [Google Scholar] · Zbl 1347.62155 |
| [17] | Stein C. Inadmissibility of usual estimator for the mean of a multivariate Normal distribution. Proceedings of the Third Berkeley Symposium on Mathematical Statistics and Probability. Berkeley: University of California Press; 1956. p. 197-206. [Google Scholar] · Zbl 0073.35602 |
| [18] | Aslam M.Performance of Kibria’s method for the heteroscedastic ridge regression model: some Monte Carlo evidence. Commun Statist Simul Comput. 2014;43:673-686. doi: 10.1080/03610918.2012.712185[Taylor & Francis Online], [Web of Science ®], [Google Scholar] · Zbl 1291.62136 |
| [19] | Kibria BMG.Performance of some new ridge regression estimators. Commun Statist Simul Comput. 2003;32:419-435. doi: 10.1081/SAC-120017499[Taylor & Francis Online], [Web of Science ®], [Google Scholar] · Zbl 1075.62588 |
| [20] | Liu K.Using Liu-type estimator to combat collinearity. Commun Statist Theory Methods. 2003;32:1009-1020. doi: 10.1081/STA-120019959[Taylor & Francis Online], [Web of Science ®], [Google Scholar] · Zbl 1107.62345 |
| [21] | Longley JW.An appraisal of least squares programs for the electronic computer from the point of view of the user. J Amer Statist Assoc. 1967;62:819-841. doi: 10.1080/01621459.1967.10500896[Taylor & Francis Online], [Web of Science ®], [Google Scholar] |
| [22] | Cook RD.Detection of influential observations in linear regression. Technometrics. 1977;19:15-18. [Taylor & Francis Online], [Web of Science ®], [Google Scholar] · Zbl 0371.62096 |
This reference list is based on information provided by the publisher or from digital mathematics libraries. Its items are heuristically matched to zbMATH identifiers and may contain data conversion errors. In some cases that data have been complemented/enhanced by data from zbMATH Open. This attempts to reflect the references listed in the original paper as accurately as possible without claiming completeness or a perfect matching.
