Monday Message Board

It’s time for the Monday Message Board, where readers are invited to post their thoughts on any topic (civilised discussion and no coarse language, please). There will doubtless be plenty of posts from me on the election, and plenty of room for discussion, so I’d encourage Message Board comments on other issues.

The markets and the election

Commenter “tipper” supplies the numbers on an issue I’ve discussed previously, observing

Centrebet has the Coalition at $1.55 and Labor at $2.30. For the non-punters, that means you have to bet $7 to win $4 on the Coalition and $10 to win $13 on Labor. I think I read somewhere recently, that the bookies have picked the elections better than the polls for the last couple of years. So go all you “true believers”, make an honest quid for yourselves, for the first time in your lives, by proving the bookies wrong. Or as John would put it, prove the “efficient market hypothesis” wrong.

If I’ve done my arithmetic properly, and allowing for the bookies’ margin, I get the implied probabilities as 0.60 for the Coalition and 0.40 for Labor. The polls have Labor ahead, but looking at all the discussion, I’d say that the consensus view is that the election is a 50-50 proposition, and that’s also my subjective probability.

How good a test of the efficient markets hypothesis will this be? Bayesian decision theory provides an answer[1]. If our initial belief is that the EMH is equally likely to be true or false, and the Coalition wins, we should revise our probability for the EMH up to 0.55. If Labor wins, we should revise it down to 0.45.

As regards the betting option, there’s a collective decision problem here. Given my subjective probabilities, a bet of $100 on Labor would have an expected net payoff of $15, but $15 isn’t enough to cover my transactions costs for placing the bet, etc. A bet of $1000 would have an expected net payoff of $150, which would be worthwhile in these terms. Unfortunately, the 50 per cent chance of losing the $1000 comes with the additional cost of having to explain to my (non-Bayesian) wife what a good choice I had made ex ante . The net expected benefit comes out as a big negative here.

fn1. The workings are easy for those who know Bayes’ theorem and accept the modern subjectivist interpretation , but they won’t make much sense to those who don’t.

The idea of the university

Before we get bogged down in the election, I thought I’d do a quick post on this piece that caught my eye. It was a piece in the Age in the long-running dispute between Melbourne University Private and Senator Kim Carr, Labor Science and Research Spokesman. Here’s what caught my eye.

[Carr] claimed five of the 12 research publications MUP had produced up to June 30 this year did not meet Commonwealth Department of Education, Science and Training guidelines and suggested there were doubts about another five.

The university, in its final submission to the committee on Thursday, denied making any false claims, maintained its research report had been properly audited, and said its research publication rate was almost twice the national average (emphasis added).

I thought this must be a misprint. An entire university with 12 publications in a year? I get more than that[1], and so do quite a few other researchers. And how can this possibly be “above the national average?”
Read More »

It’s on !

PM Calls Oct 9 poll In the face of such a long-anticipated announcement, I haven’t anything new to say right now, so I thought I’d invite commenters to post their predictions. Any sort of prediction is welcome, but I’ll have a contest for the commenter who can give the most accurate forecast of the number of House of Representatives seats won by the Coalition. The prize, if any, will be announced later.,

Misleading reporting of surveys

A lot of people have pointed to a survey reported in the Telegraph and accepted the misleading interpretation of the results given by the Tele’s Malcolm Farr. First, the results. Of those surveyed, 47 per cent thought Howard had lied on children overboard, 31 per cent did not believe he had lied, and 22 per cent were uncommitted. On a second question, 60 per cent said it would not influence the way they voted, 31 per cent said it would and 9 per cent were uncommitted. The heading is “If PM lied, few care” and the opening para is “A large majority of voters have said the issue won’t influence their polling booth choice – even if they think the Prime Minister didn’t tell the truth.” This is wrong in almost every way a survey report can be wrong.
Read More »

Howard the spammer

The right wing of Ozplogistan has lined up fairly loyally behind Howard until now. There are plenty of people happy to vote for a liar. But is anyone out there loyal enough to advocate voting for a spammer ?[1],[2]

fn1. OK, it’s Howard’s son who actually does the spamming, but it’s Howard’s money and message.

fn2. While I’m at it, a quick note on my previous post on “legitimate” companies that employ spammers I emailed two of those mentioned (couldn’t find an email address on the website for the third) but got no reply.