You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
Why is it not possible to specify two separate, and contradictory, hypotheses? Currently bain throws an error. If two experts disagree about the proper hypothesis, wouldn't Bayes factors be a great way to settle the debate?
The text was updated successfully, but these errors were encountered:
De Bayes factor moet consistent zijn. Voor een bepaalde klasse van hypotheses (comparable) hebben wij een consistente Bayes factor ontworpen. Voor, bijvoorbeeld, de vergelijking van H1: m=0 versus H2: m>0 is er momenteel geen onderzoek over hoe dan de prior moet worden gekozen. Ook is niet bekeken of de Bayes factor dan consistent is. Daar kan onder zoek naar gedaan worden. Maar de benadering zoals geimplementeerd in Bain kan dit niet aan. Vrijdag hebben we een afspraak, dan zal ik het kort uitleggen en je (als je wilt :-)) de technische achtergrond hiervoor sturen.
Why is it not possible to specify two separate, and contradictory, hypotheses? Currently bain throws an error. If two experts disagree about the proper hypothesis, wouldn't Bayes factors be a great way to settle the debate?
The text was updated successfully, but these errors were encountered: