![\"Network_advanced\"](\"http:\/\/blog.rickstafford.com\/wp-content\/uploads\/2015\/01\/Network_advanced.jpg\")
<\/a><\/p>\nAnd some new functions \u2013 quite simple, what will happen to expert opinion should UKIP votes increase (or decrease). Everything else is the same.<\/p>\n
In the network, the UKIP node is blue. The reason for that is that this can be \u2018observed\u2019. Being observed gives the node a probability of 1 (of increasing \u2013 OR of decreasing). Of course, we can\u2019t be that certain, but we can use this to help predict what might happen, depending on what happens to UKIP.<\/p>\n Fundamentally, if we don\u2019t know what happens to UKIP, then the pollsters have no idea who will win. We get the same probabilities as before (62% likely for a labour win). However, if the UKIP vote decreases (let\u2019s assume the highly unlikely event of Nigel Farage saying something offensive in the next 4 and a half months\u2026.. ) then we observe that the UKIP vote goes down and\u2026 it\u2019s now only 53% likely that Labour will get the most votes.<\/p>\n Simple Bayesian belief networks are fine for these simple interactions \u2013 but let\u2019s go back to the football. Only four teams can qualify for Europe, so if one becomes more likely, this should effect the chances of the others. Simple Bayesian networks don\u2019t cope well with this \u2013 but in what is likely to be the final part of these blogs, I\u2019ll show how this can be overcome and we can incorporate interaction, data and opinion to make some final predictions.<\/p>\n","protected":false},"excerpt":{"rendered":" In my previous post, I made the insinuation that election commentators had no idea who would win the election. Broadly this is true, but of course, when you are paid to write opinion, you have to say more than this. … Continue reading <\/a> <\/a><\/p>\n