Asked by Chimezie Umezie
on 18 Aug 2012

I want to create an ANOVA table to enable carry out model adequacy checking using Lack of Fit sum of squares and pure error sum of squares of a dataset using the

“tbl=anova(mdl,anovatype)” syntax and mdl=LinearModel.fit(X, y) where X is Matrix of predictor values and y is Vector of response values with the same number of rows as X. I have followed the syntax provided in the following links :

http://www.mathworks.com/help/toolbox/stats/linearmodel.fit.html & http://www.mathworks.com/help/toolbox/stats/linearmodel.anova.html

but I get this error message “??? Undefined variable "LinearModel" or class "LinearModel.fit".”

I even tried using the example in the above link and the message is the same. I do not know where the problem lies. I would really appreciate if someone could guide me in solving this problem. I am using Matlab version 7.5.Below is the case study.

Case study Model type: Linear Control variables are Conc and Speed and the response is the yield. Conc=[45 55 45 55 50 50 50]';Speed=[90 90 110 110 100 100 100]'; yield=[69 59 78 67 68 66 69]'; Seven runs were carried out, three of these being replicate measurements at the center point.

*No products are associated with this question.*

Answer by Oleg Komarov
on 18 Aug 2012

Edited by Oleg Komarov
on 18 Aug 2012

Accepted answer

Not available for your version. Consider browsing the documentation of your release: http://www.mathworks.co.uk/help/doc-archives.html

On R2011a you have ANOVA.

Opportunities for recent engineering grads.

## 0 Comments