Perturbation selection and influence measures in local influence analysis

Cook's [J. Roy. Statist. Soc. Ser. B 48 (1986) 133-169] local influence approach based on normal curvature is an important diagnostic tool for assessing local influence of minor perturbations to a statistical model. However, no rigorous approach has been developed to address two fundamental issues: the selection of an appropriate perturbation and the development of influence measures for objective functions at a point with a nonzero first derivative. The aim of this paper is to develop a differential-geometrical framework of a perturbation model (called the perturbation manifold) and utilize associated metric tensor and affine curvatures to resolve these issues. We will show that the metric tensor of the perturbation manifold provides important information about selecting an appropriate perturbation of a model. Moreover, we will introduce new influence measures that are applicable to objective functions at any point. Examples including linear regression models and linear mixed models are examined to demonstrate the effectiveness of using new influence measures for the identification of influential observations.

[1]  B. Efron.  Defining the Curvature of a Statistical Problem (with Applications to Second Order Efficiency) , 1975 .

[2]  S. Weisberg,et al.  Residuals and Influence in Regression , 1982 .

[3]  Shun-ichi Amari,et al.  Differential-geometrical methods in statistics , 1985 .

[4]  Peter McCullagh,et al.  Invariants and likelihood ratio statistics , 1986 .

[5]  D. Cox,et al.  Parameter Orthogonality and Approximate Conditional Inference , 1987 .

[6]  Christopher J. Nachtsheim,et al.  Diagnostics for mixed-model analysis of variance , 1987 .

[7]  Anthony J. Lawrance,et al.  Regression Transformation Diagnostics Using Local Influence , 1988 .

[8]  Chih-Ling Tsai,et al.  Transformation-model diagnostics , 1992 .

[9]  Xizhi Wu,et al.  Second‐Order Approach to Local Influence , 1993 .

[10]  Residual sum of squares and multiple potential, diagnostics by a second order local approach , 1993 .

[11]  M. Murray,et al.  Differential Geometry and Statistics , 1993 .

[12]  R. Dennis Cook,et al.  Leverage, local influence and curvature in nonlinear regression , 1993 .

[13]  J. Leventhal,et al.  Maltreatment of children born to cocaine-dependent mothers. , 1993, American journal of diseases of children.

[14]  A. Berg,et al.  Are children born to young mothers at increased risk of maltreatment? , 1993, Pediatrics.

[15]  P. McCullagh,et al.  Potential functions and conservative estimating functions , 1994 .

[16]  D. Ruppert,et al.  Measurement Error in Nonlinear Models , 1995 .

[17]  Wing K. Fung,et al.  A Note on Local Influence Based on Normal Curvature , 1997 .

[18]  B. Wei,et al.  Preferred point a-manifold and Amari's a-connections , 1997 .

[19]  R. Kass,et al.  Geometrical Foundations of Asymptotic Inference , 1997 .

[20]  Heping Zhang,et al.  Multivariate Adaptive Splines for Analysis of Longitudinal Data , 1997 .

[21]  W. Fung,et al.  Generalized Leverage and its Applications , 1998 .

[22]  Heping Zhang.  Analysis of Infant Growth Curves Using Multivariate Adaptive Splines , 1999, Biometrics.

[23]  Yat Sun Poon,et al.  Conformal normal curvature and assessment of local influence , 1999 .

[24]  W. Fung,et al.  Influence Analysis for Linear Measurement Error Models , 2000 .

[25]  M. C. Chaki.  ON STATISTICAL MANIFOLDS , 2000 .

[26]  Sik-Yum Lee,et al.  Local influence for incomplete data models , 2001 .

[27]  M. Berger,et al.  Local Influence to Detect Influential Data Structures for Generalized Linear Mixed Models , 2001, Biometrics.

[28]  P M Bentler,et al.  Effect of outliers on estimators and tests in covariance structure analysis. , 2001, The British journal of mathematical and statistical psychology.

[29]  Ana Ivelisse Avilés,et al.  Linear Mixed Models for Longitudinal Data , 2001, Technometrics.

[30]  G Molenberghs,et al.  Sensitivity Analysis for Nonrandom Dropout: A Local Influence Approach , 2001, Biometrics.

[31]  Wing K. Fung,et al.  Influence diagnostics and outlier tests for semiparametric mixed models , 2002 .

[32]  Sik-Yum Lee,et al.  Local influence for generalized linear mixed models , 2003 .

[33]  Xuming He,et al.  Local Influence Analysis for Penalized Gaussian Likelihood Estimators in Partially Linear Models , 2003 .

[34]  Andrew M. Kuhn,et al.  Growth Curve Models and Statistical Diagnostics , 2003, Technometrics.

[35]  Nils Lid Hjort,et al.  Goodness of Fit via Non‐parametric Likelihood Ratios , 2004 .

[36]  Heping Zhang,et al.  A diagnostic procedure based on local influence , 2004 .

[37]  Sik-Yum Lee,et al.  Local influence analysis of nonlinear structural equation models , 2004 .

[38]  S. Lipsitz,et al.  Missing-Data Methods for Generalized Linear Models , 2005 .

[39]  Universitext.  An Introduction to Ordinary Differential Equations , 2006 .