Following the appointment of the new Cabinet, the Forest Sector now reports to the ministère des Ressources naturelles et des Forêts, while the Wildlife and Parks Sectors report to the ministère de l'Environnement, de la Lutte aux changements climatiques, de la Faune et des Parcs. Adjustments will be made to the website over time.

Back to publications

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

Summary

Published in European Journal of Forest Research 126: 573-585

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 homegeneous variances. Moreover, compared with traditional nonlinear least sqares (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.

Sector(s): 

Forests

Categorie(s): 

Scientific Article

Theme(s): 

Forestry Research, Forests, Silviculture

Author(s)

FORTIN, Mathieu, Gaétan DAIGLE, Chhun-Huor UNG, Jean BÉGIN and Louis ARCHAMBAULT

Year of publication :

2007

Keywords :

nonlinear modeling, variance modeling, covariance structure, predictions, red spruce, silviculture and yield of natural forests - hardwood stands, silviculture and yield of natural forests - mixed stands, silviculture and yield of natural forests - softwood stands, balsam fir, Picea rubens, Abies balsamea

Partagez