A variance-covariance structure to take into account repeated measurements and heteroscedasticity in growth modeling

This study proposes a within-subject variance-covariance (VC) structure to take into account repeated measurements and heteroscedasticity in a context of growth modeling. The VC structure integrates a variance function and a continuous autoregressive covariance structure. It was tested on a nonlinear growth model parameterized with data from permanent sample plots. Using a stand-level approach, basal area growth was independently modeled for red spruce (Picea rubens Sarg.) and balsam fir [Abies balsamea (L.) Mill.] in mixed stands. For both species, the implementation of the VC structure significantly improved the maximum likelihood of the model. In both cases, it efficiently accounted for heteroscedasticity and autocorrelation, since the normalized residuals no longer exhibited departures from the assumptions of independent error terms with homogeneous variances. Moreover, compared with traditional nonlinear least squares (NLS) models, models parameterized with this VC structure may generate more accurate predictions when prior information is available. This case study demonstrates that the implementation of a VC structure may provide parameter estimates that are consistent with asymptotically unbiased variances in a context of nonlinear growth modeling using a stand-level approach. Since the variances are no longer biased, the hypothesis tests performed on the estimates are valid when the number of observations is large.

[1]  J.C.G. Goelz,et al.  Development of a well-behaved site index equation: jack pine in north central Ontario , 1992 .

[2]  R. Bailey,et al.  A Multivariate Simultaneous Prediction System for Stand Growth and Yield with Fixed and Random Effects , 2001 .

[3]  Susan A. Murphy,et al.  Monographs on statistics and applied probability , 1990 .

[4]  E. Vonesh,et al.  Mixed-effects nonlinear regression for unbalanced repeated measures. , 1992, Biometrics.

[5]  A. Gallant,et al.  Nonlinear Statistical Models , 1988 .

[6]  J. Dhôte,et al.  A flexible radial increment model for individual trees in pure even-aged stands , 2004 .

[7]  Timothy G. Gregorie,et al.  Generalized Error Structure for Forestry Yield Models , 1987, Forest Science.

[8]  Russ Wolfinger,et al.  Computing Gaussian Likelihoods and Their Derivatives for General Linear Mixed Models , 1994, SIAM J. Sci. Comput..

[9]  Timothy G. Gregoire,et al.  Model Fitting Under Patterned Heterogeneity of Variance , 1989, Forest Science.

[10]  Marie Davidian,et al.  Nonlinear Models for Repeated Measurement Data , 1995 .

[11]  Robert L. Bailey,et al.  Modeling and Prediction of Forest Growth Variables Based on Multilevel Nonlinear Mixed Models , 2001 .

[12]  Michael Clutter,et al.  Multivariate Multilevel Nonlinear Mixed Effects Models for Timber Yield Predictions , 2004, Biometrics.

[13]  H. Pretzsch,et al.  A Re-Evaluation of Reineke's Rule and Stand Density Index , 2005, Forest Science.

[14]  Valerie LeMay,et al.  MSLS: a linear least squares technique for fitting a simultaneous system of equations with a generalized error structure , 1990 .

[15]  F. J. Richards.  A Flexible Growth Function for Empirical Use , 1959 .

[16]  S. R. Searle,et al.  Generalized, Linear, and Mixed Models , 2005 .

[17]  L. Pienaar,et al.  The Chapman-Richards Generalization of Von Bertalanffy's Growth Model for Basal Area Growth and Yield in Even - Aged Stands , 1973 .

[18]  J. Hervé,et al.  Changements de productivité dans quatre forêts de chênes sessiles depuis 1930 : une approche au niveau du peuplement , 2000 .

[19]  Douglas A. Maguire,et al.  Modeling stem taper of three central Oregon species using nonlinear mixed effects models and autoregressive error structures , 2003 .

[20]  D. Ruppert,et al.  Transformation and Weighting in Regression , 1988 .

[21]  Juha Lappi,et al.  A non-linear hierarchical mixed model to describe tree height growth , 2006, European Journal of Forest Research.

[22]  M. Westveld.  Ecology and Silviculture of the Spruce-Fir Forests of Eastern North America , 1953 .

[23]  N. Duan.  Smearing Estimate: A Nonparametric Retransformation Method , 1983 .

[24]  L. Skovgaard.  NONLINEAR MODELS FOR REPEATED MEASUREMENT DATA. , 1996 .

[25]  Richard Valliant,et al.  Finite population sampling and inference : a prediction approach , 2000 .

[26]  J. L. Clutter,et al.  A Simultaneous Growth and Yield Model for Loblolly Pine , 1972 .

[27]  B. Parresol.  Modeling Multiplicative Error Variance: An Example Predicting Tree Diameter from Stump Dimensions in Baldcypress , 1993, Forest Science.

[28]  R. Littell.  SAS System for Mixed Models , 1996 .

[29]  V. Carey,et al.  Mixed-Effects Models in S and S-Plus , 2001 .

[30]  C. Ung,et al.  Relating Site Index to Ecological Factors in Black Spruce Stands: Tests of Hypotheses , 1999, Forest Science.

[31]  H. Pretzscha,et al.  The single tree-based stand simulator SILVA : construction , application and evaluation , 2002 .

[32]  D. Bates,et al.  Mixed-Effects Models in S and S-PLUS , 2001 .

[33]  Robert L. Bailey,et al.  Nonlinear Mixed Effects Modeling for Slash Pine Dominant Height Growth Following Intensive Silvicultural Treatments , 2001 .

[34]  G. J. Mitchell,et al.  Principles and procedures of statistics: A biometrical approach , 1981 .

[35]  Jerome K. Vanclay,et al.  Modelling Forest Growth and Yield: Applications to Mixed Tropical Forests , 1994 .

[36]  Timothy G. Gregoire,et al.  Linear modelling of irregularly spaced, unbalanced, longitudinal data from permanent-plot measurements , 1995 .

[37]  D. Bates,et al.  Newton-Raphson and EM Algorithms for Linear Mixed-Effects Models for Repeated-Measures Data , 1988 .

[38]  P. Krestov,et al.  Height growth of black spruce in British Columbia , 2002 .

[39]  L. Bélanger,et al.  Les coupes partielles : une alternative à la coupe à blanc dans les peuplements mixtes de sapin baumier et d'épinette rouge en termes de rendement , 2003 .