 |
 |
Thanks! I am looking forward to try it out. Best Torben Torben E. Lund Danish Research Centre for MR Copenhagen University Hospital Kettegaard Allé 30 2650 Hvidovre Denmark email: [log in to unmask] webpage: http://www.drcmr.dk On 17 May 2005, at 12:41, Will Penny wrote: > Hello Torben, > > In the model *specification* stage only error models > 'AR(1)' or 'none' can be selected. In the help text at the bottom > of the screen it says 'If you choose Bayesian estimation this > option will be ignored'. > > What it does'nt say but should ! (and I'll change it so it does !) is > 'For Bayesian > estimation AR model options are specified at the model estimation > stage'. > > So, specify your model as usual - press the run button to create your > design > matrix. Then press Estimate. Then under Method, select 'Bayesian > 1st-level'. Expand > that option and a host of sub-options will be provided, including AR > model order. > > Hope this helps. We are working on a manual and will get it to the > community > as soon as possible. > > Best, > > Will. > > Torben Ellegaard Lund wrote: > >> Dear Will and SPM'ers >> How do I model higher order serial correlations in SPM5b, it doesn't >> seem possible via the user interface? >> Best >> Torben >> Torben E. Lund >> Danish Research Centre for MR >> Copenhagen University Hospital >> Kettegaard Allé 30 >> 2650 Hvidovre >> Denmark >> email: [log in to unmask] >> webpage: http://www.drcmr.dk > > -- > William D. Penny > Wellcome Department of Imaging Neuroscience > University College London > 12 Queen Square > London WC1N 3BG > > Tel: 020 7833 7475 > FAX: 020 7813 1420 > Email: [log in to unmask] > URL: http://www.fil.ion.ucl.ac.uk/~wpenny/