Message-ID: <D3557409FD94BA41AE028A1E573716D646FBAFBA@S-BSC-MBX2.nrn.nrcan.gc.ca>
Date: 2013-04-30T17:21:32Z
From: Boulanger, Yan
Subject: Convergence problem
In-Reply-To: <1367277115.23904.YahooMailNeo@web160504.mail.bf1.yahoo.com>
Hi folks,
I'm new to nlme and I'm trying to fit a somewhat simple mixed regression model:
PICE.MAR_tree_dbh_model <- lme(log(dbh_tree) ~ Temp+Prec+I(Prec/age_tree)+I(Temp/age_tree), weights = varPower(form =~fitted(.)) , random= ~1|plot_name, data=PICE.MAR_tree, method="REML")
However, convergence is never reached:
(Error in lme.formula(log(dbh_tree) ~ Temp + Prec + I(Prec/age_tree) + :
maximum number of iterations (lmeControl(maxIter)) reached without convergence)
Here's the structure of my data. dbh_tree and age_tree is measured on the tree which is nested in plot_name (random effect). Temp and Prec are measured at the plot level. I have also a very unbalanced setup with sometimes only 1 tree per plot and others where there are as much as 150 trees. Maybe the syntax is not right.
Thank you!
Yan Boulanger
>Yan Boulanger, Postdoctoral Visiting Fellow
>Ressources Naturelles Canada, Canadian Forest Service
>Centre de Foresterie des Laurentides
>1055, rue du P.E.P.S.
>C.P. 10380, succ. Sainte-Foy
>Qu?bec (Qu?bec) Canada
>G1V 4C7
>Tel. : +1 418 649-6859