I had problem in estimating a DCM model on one subject ( Matrix is close to singular or badly scaled). But the same model can be estimated on many other subjects. I wonder what is the problem. Is it due to problem in model or this particular subject. How should I fix this problem? I really appreciated if anyone can help.
SPM gave following error message:
SPM8: spm_dcm_estimate: model 1 19:38:28 - 26/03/2011
----------------------------------------------------------------
EM:(+): 1 F: 0.000e+00 dF predicted: 9.501e+01 actual: 5.422e+01 (1122.44 sec)
EM:(+): 2 F: 5.422e+01 dF predicted: 2.243e+02 Warning: Matrix is close to singular or badly scaled.
Results may be inaccurate. RCOND = -1.000000e+00.
> In spm_inv at 22
In spm_nlsi_GN at 279
In spm_dcm_estimate at 172
In spm_dcm_ui at 126
Warning: Matrix is singular, close to singular or badly scaled.
Results may be inaccurate. RCOND = NaN.
> In spm_expm at 51
In spm_dx at 83
In spm_nlsi_GN at 311
In spm_dcm_estimate at 172
In spm_dcm_ui at 126
Warning: Matrix is singular, close to singular or badly scaled.
Results may be inaccurate. RCOND = NaN.
> In spm_nlsi_GN at 277
In spm_dcm_estimate at 172
In spm_dcm_ui at 126
Warning: Matrix is close to singular or badly scaled.
Results may be inaccurate. RCOND = -1.000000e+00.
Best
