Help with Linear mixed effects Model (LME)

1 次查看（过去 30 天）
Luís Lamas
Luís Lamas 2018-11-8
回答： Paul 2020-8-5
Hi! I have a complicated dataset of biomechanical analysis where I am looking at different joint angles and the way they change over time (in % of of cycle) in different conditions (2 types). I need to investigate the influence of subjects (3 of them) and timepoint (1 to 101) on the data. I want to regress angle on condition.
I have prepared an a table with the variables for lme.
lme = fitlme(tbl,'Angle~Condition+(1|Subject)+(Conditon-1|Subject)');
Condition and Subject are categorical data on tbl.
When I run this line I keep getting the error message:_
"Class 'classreg.regr.LinearLikeMixedModel' is not an allowed subclass of class 'classreg.regr.ParametricRegression'.
Error in fitlme (line 224) lme = LinearMixedModel.fit(ds,formula,varargin{:});"
Any help on modelling this and why I am getting this error?
Thanks

请先登录，再回答此问题。

回答（1 个）

Paul
Paul 2020-8-5
Hi Luís,
That's a rather unfamiliar error to me. Could you provide the code you use to produce the table and some portion of the contents of the table so I might reproduce the error?
Also, how many datapoints (rows) are in your table?
And did all three subjects provide data for both conditions?
Also, I believe your formula can be simplified to 'Angle ~ Condition + (Condition|Subject)'. Is there any particular reason you have it formulated the way you did?
Best,
Paul

类别

在 Help Center 和 File Exchange 中查找有关 Programming 的更多信息

标签

产品


版本

R2015b

Community Treasure Hunt

Find the treasures in MATLAB Central and discover how the community can help you!

Start Hunting!

Translated by