problem with the anova calculation

I have a problem with the calculation of the ANOVA. What is wrong with my code?

C_t = [30; 30; 30; 30; 30; 30; 30; 120; 120; 120; 120; 120; 120; 120; 120];
pick = [0; 0.026;  0.0550;  0.120 ; 0.320 ; 0.62 ; 1.2; 0 ;0.03;  0.066;  0.110 ; 0.220;  0.4;  0.740;  1.73];
tbl = table(C_t,pick,'VariableNames',{'C_t','pick'});
tbl.C_t = categorical(tbl.C_t);
mdl = fitlm(tbl,'pick')
tbl = anova(mdl,'summary'

Answer 1

[Jeff Miller]

Just change one line to this

mdl = fitlm(tbl,'pick~C_t')

[Michael]

Thank you very much! Now it works. But I have one more question: why doesn't the function tbl = anova(mdl,'summary') work as in matlab help?

tbl = anova(mdl,'summary')

tbl=7×5 table

SumSq DF MeanSq F pValue

______ __ ______ _______ ________

Total 4757.8 99 48.059

Model 274.73 3 91.577 1.961 0.12501

. Linear 243.8 2 121.9 2.6103 0.078726

. Nonlinear 30.934 1 30.934 0.66242 0.41772

Residual 4483.1 96 46.699

. Lack of fit 1483.1 39 38.028 0.72253 0.85732

. Pure error 3000 57 52.632

I don't have (Liner, nonlineral, Lack of fit and pure error) and I care about these values

[Jeff Miller]

You don't have linear, etc, because you have indicated that C_t is categorical. The missing terms are only calculated for numerical predictors.

But the example data you posted have only 2 values for C_t. It is not possible to distinguish linear versus nonlinear trends across only two C_t values.

[Michael]

Ok, but how can i calculate leak of fit and purre error from this data?

[Jeff Miller]

You can't. This is not a software problem, it is a limitation of your design. With only two values of X (30, 120), a straight line goes through the two (X, avg Y) pairs perfectly, so your data provide no information about whether there would be any deviation from a straight line fit ("lack of fit") if you had a third X value.

Answer 2

[James Water Bird]

C_t = [30; 30; 30; 30; 30; 30; 30; 120; 120; 120; 120; 120; 120; 120; 120];
pick = [0; 0.026;  0.0550;  0.120 ; 0.320 ; 0.62 ; 1.2; 0 ;0.03;  0.066;  0.110 ; 0.220;  0.4;  0.740;  1.73];

% Use the table() function to create a table from the data
tbl = table(C_t,pick,'VariableNames',{'C_t','pick'});

% Convert the 'C_t' column to a categorical variable
tbl.C_t = categorical(tbl.C_t);

% Use the fitlm() function to fit a linear regression model to the data
mdl = fitlm(tbl,'pick')

% Use the anova() function to perform an analysis of variance on the model
tbl = anova(mdl,'summary')

[Michael]

I understand that you have added descriptions, but this does not solve my problem.