I am Getting a NaN when using ANOVA function for regression.

I am using the function anova for feature selection for regression. My model has 29 features. I performed zscore standardization on all the features and the targetted output before using the function. I made sure that there are no NaN values within the data set. I used fitlm to generate the model then fed it into the anova functon.

This the the resultant ANOVA table:

SumSq DF MeanSq F pValue

_________ _____ _________ _________ __________

x1 348.22 1 348.22 532.3 3.756e-116

x2 41.266 1 41.266 63.082 2.095e-15

x3 30.912 1 30.912 47.253 6.4354e-12

x4 0.53704 1 0.53704 0.82095 0.36491

x5 0.04193 1 0.04193 0.064096 0.80014

x6 2.6268 1 2.6268 4.0155 0.045098

x7 1.9056 1 1.9056 2.913 0.087887

x8 242.84 1 242.84 371.22 6.3464e-82

x9 1.6373 1 1.6373 2.5028 0.11366

x10 61.105 1 61.105 93.408 4.7937e-22

x11 0 0 NaN NaN NaN

x12 0 0 NaN NaN NaN

x13 0 0 NaN NaN NaN

x14 0 0 NaN NaN NaN

x15 40.628 1 40.628 62.106 3.4331e-15

x16 10.361 1 10.361 15.839 6.923e-05

x17 42.666 1 42.666 65.222 7.0958e-16

x18 6.8152 1 6.8152 10.418 0.00125

x19 2.4515 1 2.4515 3.7475 0.052902

x20 0.20204 1 0.20204 0.30885 0.57839

x21 2.6684 1 2.6684 4.0791 0.043431

x22 0.027859 1 0.027859 0.042587 0.83651

x23 10.786 1 10.786 16.489 4.9139e-05

x24 4.289 1 4.289 6.5564 0.010458

x25 0.0048784 1 0.0048784 0.0074574 0.93118

x26 6.4326 1 6.4326 9.8333 0.0017165

x27 30.219 1 30.219 46.195 1.1029e-11

x28 221.09 1 221.09 337.97 8.0981e-75

x29 1237.3 1 1237.3 1891.3 0

Error 12201 18651 0.65417

[Dheeraj Singh]

You can have a look at the following MATLAB Answer:

https://in.mathworks.com/matlabcentral/answers/102853-why-do-i-recieve-nan-s-in-the-anova-table-when-i-try-to-analyze-my-data-using-the-statistics-toolbox