ttest returns NaN even though matrices are finite and does not contain nan?

I have 2 matrices: sample1 - 17*80 and sample2 - 17*80

find(isnan(sample1)==1) and find(isnan(sample2)==1) are empty matrices.

Similarly, find(isfinite(sample1)==0) and find(isfinite(sample2)==0) are empty matrices.

[h,p] = ttest(sample1, sample2) results into a 1*80 vector, each value = NaN.

Why does it happen so??

[Katerina Merculy]

I have the same problem and vextors are not the same, they even have different number of rows and I compare ttest(mean(sample1),mean(sample2)) Have you decided this question?

[Stephen23]

@Katerina Merculy: please upload your data in a .mat file.

Answer 1

[Anthony Peng]

Are the two vectors the same? If they have the same values in them, it will return NaN.

[Katerina Merculy]

I have the same problem and vectors are not the same, they even have different number of rows and I compare ttest(mean(sample1),mean(sample2))

[ANKUSH JAMTHIKAR]

I have two binary vectors with same elements. I mean, if X and Y are the two binary vectors, then X=Y. Now when I am running ttest(X,Y), it is giving me h=NaN and p=NaN. Does anyone know, how to handel this issue?

I am a bigginner in statistics. So kindly help me to understand this.

Thank you in advanced!

[Adam Danz]

@Katerina Merculy, you're getting NaN values because mean(sample1) and mean(sample2) result in scalar values which is interpretted as a 1-sample ttest (not a paired ttest) and you need at least 2 values to carry out the test. Instead of inputting the mean, input the raw values.

@ANKUSH JAMTHIKAR, when the x and y inputs to ttest(x,y) are exactly equal, you should expect a NaN results.

For more detail on both problems, see this answer which explains in detail why NaN results occur in ttest().