Hi Anderson,

I put the original values without rounding off. Since I could not paste the copied data, I entered them manually.
Anyway, I restarted randomization on a same computer but it does not work after showing the warning: "tfce has detected a large number of integral steps. This operation may require a great deal of time to complete".
Even though 14 hours later, the calculation have not progressed at all.
Is it because I put the complicated values (with the order of 10^-16)?

Here I attached the matrices just in case.


Best regards,
Akira Yogi

On Tue, 4 Aug 2015 07:44:37 +0100, Anderson M. Winkler <[log in to unmask]> wrote:

>Hi Akira,
>
>Excel can give values very close to zero (something to the order of 10^-16
>I think). Check the formatting options as these other decimal places may be
>hidden. Also, use copy/paste to put the files in the Glm_gui window, so
>that these are preserved.
>
>All the best,
>
>Anderson
>
>
>
>On 4 August 2015 at 00:58, Akira Yogi <[log in to unmask]> wrote:
>
>> Hi Anderson,
>>
>> Thank you for your reply.
>> As you told, the covariate values were rounded off to two decimal places
>> by Excel's function, resulting in incomplete zero mean.
>> I will try to put the value particularly as possible as Excel can show,
>> but it seems to be impossible to have these values being completely
>> zero-mean.
>>
>>
>> Akira Yogi
>>
>