TY - JOUR
T1 - Model selection in electromagnetic source analysis with an application to VEFs
AU - Waldorp, Lourens J.
AU - Huizenga, Hilde M.
AU - Grasman, Raoul P.P.P.
AU - Böcker, Koen B.E.
AU - De Munck, Jan C.
AU - Molenaar, Peter C.M.
N1 - Funding Information:
Manuscript received December 21, 2001; revised May 1, 2002. The work of L. J. Waldorp (527-25-013), R. P. P. P. Grasman (527-25-014), and K. B. E. Böcker (527-25-015) was supported by a grant from the Netherlands Organization for Scientific Research (NWO), Foundation for Behavioral and Educational Sciences awarded to H. M. Huizenga, P. C. M. Molenaar and J. L. Kenemans. The work of H. M. Huizenga was also supported by a KNAW fellowship. Asterisk indicates corresponding author. *L. J. Waldorp is with the Department of Psychology, University of Amsterdam, Roetersstraat 15, 1018 WB Amsterdam, The Netherlands (e-mail: [email protected]).
PY - 2002/10/1
Y1 - 2002/10/1
N2 - In electromagnetic source analysis, it is necessary to determine how many sources are required to describe the electroencephalogram or magnetoencephalogram adequately. Model selection procedures (MSPs) or goodness of fit procedures give an estimate of the required number of sources. Existing and new MSPs are evaluated in different source and noise settings: two sources which are close or distant and noise which is uncorrelated or correlated. The commonly used MSP residual variance is seen to be ineffective, that is it often selects too many sources. Alternatives like the adjusted Hotelling's test, Bayes information criterion and the Wald test on source amplitudes are seen to be effective. The adjusted Hotelling's test is recommended if a conservative approach is taken and MSPs such as Bayes information criterion or the Wald test on source amplitudes are recommended if a more liberal approach is desirable. The MSPs are applied to empirical data (visual evoked fields).
AB - In electromagnetic source analysis, it is necessary to determine how many sources are required to describe the electroencephalogram or magnetoencephalogram adequately. Model selection procedures (MSPs) or goodness of fit procedures give an estimate of the required number of sources. Existing and new MSPs are evaluated in different source and noise settings: two sources which are close or distant and noise which is uncorrelated or correlated. The commonly used MSP residual variance is seen to be ineffective, that is it often selects too many sources. Alternatives like the adjusted Hotelling's test, Bayes information criterion and the Wald test on source amplitudes are seen to be effective. The adjusted Hotelling's test is recommended if a conservative approach is taken and MSPs such as Bayes information criterion or the Wald test on source amplitudes are recommended if a more liberal approach is desirable. The MSPs are applied to empirical data (visual evoked fields).
UR - http://www.scopus.com/inward/record.url?scp=0036785213&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=0036785213&partnerID=8YFLogxK
U2 - 10.1109/TBME.2002.803543
DO - 10.1109/TBME.2002.803543
M3 - Article
C2 - 12374336
AN - SCOPUS:0036785213
SN - 0018-9294
VL - 49
SP - 1121
EP - 1129
JO - IEEE Transactions on Biomedical Engineering
JF - IEEE Transactions on Biomedical Engineering
IS - 10
ER -