[R-sig-ME] call for help
carlosfamilia at gmail.com
Mon Oct 10 16:54:18 CEST 2016
I normally use the following code to check for assumptions:
plot(fitted(mdl),residuals(mdl), pch = '.')
abline(lm(y ~ x, data = data.frame(y = residuals(mdl), x = fitted(mdl))), col = 'red')
abline(a=0, b=0, lty=3)
plot(fitted(mdl), amps.long$mic, pch = '.', ylab = 'observed')
abline(lm(y ~ x, data = data.frame(y = amps.long$mic, x = fitted(mdl))), col = 'red')
abline(a = 0, b = 1, lty=3)
hist(residuals(mdl),breaks = 50, probability = T)
lines(density(residuals(mdl)), col = "red")
There are other ways to do it, and you should know what are the assumptions, for that I suggest for you to take a loot at Douglas Bates book.
Carlos Familia, MPharm, PharmD
ISCSEM Assistant Lecturer
ISCSEM Laboratory of Molecular Pathology Researcher
ISCSEM Erasmus Mundus Master in Forensic Sciences Vice-coordinator
Tel. +351 212946700 (Main) / +351 212946769 (Lab)
e-mail: carlosfamilia at gmail.com / carlosfamilia at egasmoniz.edu.pt
The information in this e-mail and any attachments may be confidential. If you have received this email in error please notify the sender immediately and remove it from your system. Do not disclose the contents to another person or take copies.
> On 10 Oct 2016, at 00:37, 罗波 <luob041 at nenu.edu.cn> wrote:
> Dear Experts,
> Nice to meet you! I am a Ph.D. student from Northeast Normal University in China. My research field is behavioral ecology. Thanks for your contribution to develop the package 'lme4'.
> That's really nice and valuable!
> Recently, I would like to fit the linear mixed model using the experimental data. Following the examples listed in the guide, I know how to achieve it using your package. Yet, I wonder if my data
> satisfy the model assumption. I have two major confusions (see followings). Would you please give me some help?
> Does it require normality of model residuals, homoscedasticity, absenceof collinearity between predictors or perhaps others?
> If so, how can I check whether my data satisfy the assumption? Would you please provide some advice or R code?
> Really need your help! Thanks very much!
> [[alternative HTML version deleted]]
> R-sig-mixed-models at r-project.org mailing list
More information about the R-sig-mixed-models