Tuesday, 20 December 2005
More phone photos
I've had to use my phone since my camera got nicked. The constraints of a tiny image size can be liberating. These are mostly of the Colne estuary, which I walk along regularly. Actually the dog is here too.
Thursday, 8 December 2005
If you believe in expert judgment
Monday, 5 December 2005
Wednesday, 16 November 2005
planet Iraq
There are getting to be so many of these people that a single page can't really do justice. Which is nice. If only all the news from Iraq were so cheerful.
Friday, 11 November 2005
just one more thing...
(Update: you should also check out the Prof's website. For a Columbia economics professor, it's pretty funny.)
More on income
The basic picture is straightforward: inequality rose markedly under Thatch. From 1979 to 1995/6, the proportion of people below half average income more or less doubled.
Of course, partly that is because average incomes rose markedly (by about 40% adjusted for inflation). But the appendices have information on the percentages of groups below various fractions of 1979 average income, held constant. In 1979, 8% of people were living below half 1979 average income. In 1995/6, 5% of people were living below half 1979 average income (adjusted for inflation). The figures for 60% of 1979 average income are similar: 18% and 10% respectively.
However, the AHC figures are much less cheerful, with the proportions of individuals living below these thresholds remaining almost constant. And the very poor do badly. There were about as many people below 40% of 1979 average income in 1995/6 as there were in 1979, whether Before or After Housing Costs.
End of brief statistical lecture. Everything is much more complex than this - for example, what about the experience of different groups? Families with children? Pensioners? What about the persistence of inequality over time? What's the right way to define or even conceptualise poverty? Despite being a Sinister Rightwinger, I believe that relative poverty is very important. Being poor, and specifically being poorer than other people, is horrible. Of course absolute poverty is bad too. Oddly enough these ideas tend to be backed up by the (stereotypically, right-wing) Darwinian approach to society: beyond satisfaction of our basic needs, we care about our relative position rather than our absolute wealth.
Corrections from the people at ISER (who do this kind of thing seriously) are of course welcome. Perhaps I should now stop reading DSS statistics on Friday night.
Just one further thought. The distribution of "talent" of many kinds is probably normal, like most things that rely on many different factors (if you add a lot of random variables together, you approach a normal distribution - the famous bell curve). The distribution of income is not normal. It looks much more like a lognormal distribution, such as the lefthand picture below
- pictures filched from http://mathworld.wolfram.com/LogNormalDistribution.html
Why? I am sure there are many extant answers to this question. My guess, keeping with the Darwinian theme, is that overachievers tend both to hang out with each other, and to compete with each other for income (or various correlates of income). This makes for a "long tail" distribution, with the mobile phone salespersons, the city slickers, and Bill Gates at the very, very top.
Seal and hard drive
Slashdot has a discussion on how long it would take for the police to decrypt an encrypted hard drive. This was one of the justifications offered for 90-day detention. Bottom line: 90 days is a slight underestimate. Good encryption cannot be cracked within the lifespan of the known universe, unless you can guess the password. Encryption is one of the subtlest parts of computer science, about which I know little, but the fundamental deal is fairly straightforward: every time you add a character (say just a-z) to your password, you multiply the number of potential passwords, and the time it takes to guess the password by brute force, by 26. Go type "26 x" into your calculator and hit "=" a few times. See?
Thursday, 10 November 2005
UK wages
Patterns of Pay 1998-2004
In particular, check out Figure 7 - Blogger seems to mess up the title:
This shows that the wages of even the lowest decile beat inflation consistently. Some credit due to New Labour? (But note that this is only for full-time employees. The pdf above has more details on the distribution of hours.)
The NES doesn't seem to have a similar time series available, but here is a useful webpage on historical average earnings. Clicking the link will show you real and nominal earnings, as well as the RPI, for 1979-2004. Really, we would prefer median earnings - they explain how they calculate average earnings but it seems complex, perhaps because they provide statistics going back to 1264!
That's all I can manage for the moment. Clearly average wages have consistently increased since 1979. The more interesting question is what has happened to the different percentiles.
Wednesday, 9 November 2005
Blair defeated over 90 days!
Update:
Now everyone is asking whether this is the end of Blair's authority. I guess it depends on the causes of the vote. Perhaps, as a premier nearing the end of his shelf life, Blair no longer wields the power to threaten backbenchers. If so, expect more revolts. Can Brown step in to quell the rebels? That probably depends on his perceived chances of winning the next election. Major's experience shows that as a government's time in office draws to a close, discipline crumbles. If that in turn makes the government appear less popular and competent, you get a vicious circle.
(Polisci geek note: I wonder if anyone has written about the game theory of the UK parliamentary system. There's lots about the US and lots about coalitions in proportional representation systems, but I don't know of any stuff for our particular setup, beyond Bagehot's famous analysis in the 19th century.)
The latest initiative - standardised testing for tiny tots - isn't exactly an inspiring big idea. More like a New Labour self-parody. If they can't do better than that I think we can expect trouble for Labour in 2009.
Monday, 7 November 2005
I'm 30
Anyway I had a party to celebrate the alleged event this weekend, which was excellent. Dave Padua, Clem, Emily Comyn and Kemal all came down from Outside. So the next day we sat in the Rose and Crown and continued to drink. Kristi put up very kindly with having a bunch of randoms on the sitting room floor, and came out with us the next day. Kemal got me a Prodigy CD. Charley says, never go out without telling your mother first. Mreowww! Clem got me the best. Card. Ever. Probably not safe to link here.
Cometh Monday, cometh the payoff....
Friday, 4 November 2005
Tuesday, 1 November 2005
iPod Nano
A present from the bruvs in America. You can see me on campus wandering around like the guy in the adverts, nodding my head in a hip and groovy way.
Epitonic have a really, really good selection of free music. I mean, St. Etienne.
A comment on ancient (direct) democracy
Working paper available
Wednesday, 19 October 2005
Seal
Tuesday, 18 October 2005
Broadcasting House
Friday, 14 October 2005
Dog news
I can speak German, honest!
Es war sehr nett, von Dir zu hoeren aber ich habe kein Email fuer Dich! Also email mir mal an. (Ist das das Wort?)
Dave
Tuesday, 27 September 2005
sloes
By Christmas or so I'll have some rich red booze. The neighbours (Pete and Sam from IDA) are getting some gin in which is the more traditional base liquor.
Wednesday, 21 September 2005
meaningless computer error messages: a collection.
Firefox. This pops up while browsing print.google.com. What document contains no data? What sort of data should it contain? What do you mean?
Copying files in Gnome. What's an invalid parameter?
This one popped up when I tried to connect to a remote server without plugging the network card in. God knows why it suddenly thinks the server is a "file" or a "folder" and starts worrying about security risks.
Tuesday, 20 September 2005
Nice weekend in Paris
Saturday night I see Viv and meet some friends of hers including the vivacious and mega-successful Elena who is an ex-World Banker. I miss the metro home and end up walking for hours through the banlieux of Northern Paris to the hotel. Yikes. Sunday we see Notre Dame and meet Viv again in a caff in front of the Sorbonne. Monday, back, another whopping trip but it gives me a chance to almost finish Shirer's Rise and Fall of the Third Reich. Shirer was a journalist, actually in Germany for a lot of the thirties, with a crackling prose style. The theme of the book is the cowardice of the people around Hitler - both nationally and internationally - and insofar as there is an explanation, at least for the Germans such as the generals who knew they were being led to destruction, he pins the blame on an inadequate political understanding. By understanding I don't just mean political science understanding, as in predicting who will do what, but also understanding of, say, the duties of a citizen or the dignity of man. (And actually these failures then lead the Nazis to make inaccurate predictions also: they cannot understand why, for example, Britain fights on rather than surrendering in 1940. It's a strong example for those who say that in social science, prediction requires understanding, and who therefore make a clear distinction between social and natural science.)
Perhaps Shirer's is a very American approach, but it basically chimes with the attitude of, say, Arendt, and contrasts strongly with the critical theorists who see fascism as the final expression of a social system - capitalism - and hence find rather little comfort in the triumph of the US. I think in this point the "politics" approach is more accurate.
As an aside: in history, I always gain insight from reading the "great" classic texts: professional historians' latest appraisals will change with the whims of academic fashion. Here at least, the judgment provided by an intelligent individual is more important than method, which may improve as the discipline progresses.
Tuesday, 13 September 2005
moved
Thursday, 8 September 2005
Cheap talk part 3
I'm going to plunge straight into this. Parts one and two are also available if you are lost.
[ed: as you can see, I end up failing to prove what I meant to prove... also, blogger keeps eating my post.]
Suppose that c = (1-a)/n where n is an integer and 0 <= a < c. In other words, there are n intervals around the unit circle, plus a remainder of a.
(1) Suppose that n is even, and that Senders of type t send a message m = t* =
t modulo c, when t* < a. Now it's a best response for Receiver to randomize equally over possible values of t in response to this message. Proof:
Let's call the possible values of t, t0 to tn where ti = ci + t*. Take any arbitrary ti. Choose an interval H = (ti, ti+1/2] or (ti, ti-1/2], such that H does not contain the zero point, and therefore does not contain both t0 and tn. (For example, from t0, go clockwise and from tn go anticlockwise.)
The interval H now contains n/2 possible values of t. Proof: ts are evenly spaced along the interval, c apart, with the first one at ti+c or ti-c (as the interval is open at ti itself). Temporarily normalizing ti to zero, the interval (0, 1/2] or (0, -1/2] is of length 1/2 and, as 1 = nc+a, 1/2 = (n/2)c + a/2 where 0 <= a/2 < c. Moving from ti into H therefore moves you towards n/2 equally likely values of t, and away from n/2+1 equally likely values of t. As we are assuming no risk-aversion (Receiver's utility is linear in distance from t), moving away from ti therefore reduces utility. Similarly, moving away in the opposite direction is moving into an interval of size 1/2, excluding ti itself, which also must contain the remaining n/2 possible values of t from the set tj, j ^= i, and away from n/2+1 equally likely values of t.
(Apologies for the HTML notation: ^= means "does not equal".)
Therefore, any a ^= ti, for all i, is strictly dominated by some pure strategy a = ti for some i, and any mixed strategies containing anything other than the tis is strictly dominated by a mixed strategy containing only tis.
Furthermore, expected distance from t is the same at any ti (leaving the proof for now, I am fairly sure) and thus the mixed strategy choosing ti with probability 1/(n+1) is a best strategy.
(2) Given this strategy, Sender's strategy is a best response. Proof: for t = ti, i e {1,2,...,n-1}, the proof is identical to that given above as Sender's ideal point is just the same as to Receiver's ideal point when t=ti+1. When t = tn, ...
ah. No wait. When t = tn Sender's ideal point will be t+c which is strictly greater than t0 = t+a (we are crossing the zero point). If so, this will surely generate an incentive to deceive Receiver by claiming to be a different type. Blast. Perhaps we can remove the interval [nc, 1) from the set of types that sends an informative message? Well, this is obviously not over yet. More pointless fun awaits.
Wednesday, 7 September 2005
Time to ditch that Yahoo! account?
If you are still using Yahoo!, why not try Gmail, Google's free email service. It's extremely easy to use and lets you import contacts from Yahoo!. Google has been accused of censoring its search engine listings for China, but at least it hasn't actually turned into a police snitch. I have several accounts to give away if anyone wants one.
Tuesday, 6 September 2005
Monday, 5 September 2005
car found
The police, of course, weren't interested in investigating. I don't really expect them to care, but it would be nice if they just pretended, you know? Of the several crimes I know in which I or friends of mine have been victims, not one has been cleared up by the police.
In other news, I just interrupted a couple of guys cottaging in the level 4 loos. Scandalous!
Sunday, 4 September 2005
My car got nicked
That has to be the least profitable car theft ever, man. The value of that car was probably negative.
Eventually we persuaded the police to take the crime report, despite them trying to persuade me that the car wasn't registered in my name on their database (it was). I don't expect them to catch the criminals or even recover the car, I just wanted them at least to acknowledge that a crime had happened.
Another amazing Katrina-related website
This is a report from the New Orleans' Times-Picayune on the potential for New Orleans to experience a devastating hurricane. Written in 2002.
Saturday, 3 September 2005
Friday, 2 September 2005
So, back in Essex...
This one was going to be written on my trendy new PDA phone, the Orange SPV M500. However, the trendy PDA phone is going back to Orange on Monday. I'd never used a PDA before and wanted to see if it's worth the hassle. Nope. Handwriting technology is way too inaccurate, and to do almost anything you need to take the stylus out and tap the screen. This is infuriating for phone features when you just want to hit a button. In general, I think I can see why PDAs have not caught on. The user interface is like an attempt to shoehorn a Windows-style computer into a phone. This just doesn't work. It's far too fiddly and over-complicated. The whole apparatus of scrollbars, popup buttons, different windows, right-clicking etc. is just not suitable. I should have stayed a loyal Nokia customer. Their basic phones have a really simple and elegant user interface, everything is natural. The apps, like the calendar, which holds my entire social life (no jokes), are just more suitable given the screen real estate.
Apart from that... well, the only other thing I wanted it for, apart from quick notes, was to blog. But if I login to blogger, Pocket Internet Explorer crashes. Bah.
why the internet is good
Any attempt to flag down police results in being told to get away at gunpoint. Hour after hour they watch buses pass by filled with people from other areas. Tensions are very high, and there has been at least one murder and several fights. 8 or 9 dead people have been stored in a freezer in the area, and 2 of these dead people are kids.
The people are so desperate that they're doing anything they can think of to impress the authorities enough to bring some buses. These things include standing in single file lines with the eldery in front, women and children next; sweeping up the area and cleaning the windows and anything else that would show the people are not barbarians.
The buses never stop.
Wednesday, 31 August 2005
Frustration, and more about the jolly game
I'm on Linux, as if you didn't know. Bah. I'm sick of having an operating system built by volunteers. I want a professional product that just works. My idealism has run out.
Now, on to more amusing things. That cheap-talk game. (You don't find abstract game theory interesting? Shame on you.)
Although it is true that there is no equilibrium in which types within different intervals along the circle play different messages, there may be an equilibrium with an infinity of messages. For example, suppose that c = 1/3 (recall that c is the difference between Sender's and Receiver's ideal points on the circle). Then, there is an equilibrium in which the message space M is [0, 1/3) and Senders send a message equal to their type, modulo 1/3. For example, senders with types 1/6, 1/2 and 5/6 would all send m = 1/6.
Receiver's best response is then any mixed strategy having as its support only actions with the possible values of Sender's type t. For example, if m = 1/6, Receiver can play any combination of 1/6, 1/2 and 5/6. The expected distance from sender's type of any one of these actions is (1/3 * 0 + 1/3*1/3 + 1/3*1/3) = 2/9. The expected distance from sender's type of any other action a is greater than this. The simplest way to see this is to imagine the three possible values of t. By moving Receiver's action a away from one of these values, two of the values get farther away, while one of them gets closer:
Now we consider Sender's best response to Receiver's strategy. First note that if Receiver plays, say "2/3" with probability 1 on receiving m = 0, but randomizes over all 3 possible types for e.g. m = 1/10, then when t = 1/3+1/10 it will be more sensible to send m = 0 (guaranteeing a response close to her bliss point) than the assumed strategy of m = 1/10. So we focus on the case when Receiver randomizes with equal probability over all 3 types, for all messages m in [0,1/3).
In this case, the logic above also holds true for Sender's utility. By sending a message reflecting her true type, she elicits a best response of one of her three possible types. The logic for Sender is just as for Receiver: this has a one in three chance of being at her bliss point (at t+c, where t+c is one of the three possible types) and a 2/3 chance of being 1/3 away; any alternative message would generate a greater expected distance from t+c. Thus, honesty is a best response to Receiver's strategy, and we have our Bayesian Nash equilibrium.
The same logic holds whenever c = 1/n, and n is odd. Then, any message m in [0,1/n) where m corresponds to t modulo 1/n will generate a best response of choosing one of the possible types: any move away from a possible type moves (n+1)/2 of the possible types away from you, and (n-1) towards you, all by the same distance.
On the other hand, when n is even, the equilibrium remains technically but is much less plausible. Now all Receiver's strategies are equally good, because moving action a away from a possible type brings n/2 possible types closer while n/2 move farther away by the same distance. The case with n=4 is shown below:
The equilibrium remains, technically, because choosing a equal to some possible value of t is still a best response. But there would be no incentive, in the long term, for Receiver to play this equilibrium. When n is odd, there are plenty of best responses that don't lead to equilibrium (such as mixed strategies where Receiver doesn't always randomize equally between points) but in the long term one might expect Receiver to play "nicely" in order to keep the flow of useful information going. (I don't know if anyone has toyed with repeated cheap-talk games.)
Before moving on to the general case, where c does not "fit neatly" into the circle, it is worth pointing out that here, as c decreases, the potential equilibria are getting less informative rather than more, in the following two senses:
- The message space for e.g. c = 1/5 is [0,1/5), which is a proper subset of the message space for c = 1/3, [0,1/3).
- The expected distance from Receiver's (and Sender's) ideal point, when c=1/n, is (n^2-1)/4(n^2). This increases as n increases, with a limit of 1/4. For example, expected distance when c = 1/3 is just 2/9, when c = 1/5 it is 6/25.
That's all for now. Next post, I'll have a shot at the general case. I suspect this will also have an equilibrium of some kind.
Monday, 29 August 2005
A jolly little cheap-talk game
The standard cheap-talk game, found e.g. in Morrow, has Sender's type t drawn from a uniform distribution on a unit interval: t ~ U[0,1]. Sender sends a message m from some set of possible messages M; then Receiver chooses an action a from that same unit interval. Receiver's utility declines monotonically, continuously and symmetrically with the distance of a from t, e.g. UR(a) = - (t-a)^2. In effect Receiver is trying to guess Sender's type. Sender's utility declines monotonically, continuously and symmetrically with distance of a from some point t+c, where c>0 and measures some conflict of interest. For example, Sender's utiliy might be US(a) = - (t+c-a)^2. So Sender wants Receiver to guess Sender's type, but be a little bit off.
It then turns out that if c is small enough, there are informative equilibria where different ranges of types send different messages (so that Receiver can infer something about Sender's type); and as c shrinks the maximum number of messages in a potential equilibrium increases. It's a nice prediction - communication gets easier when there are common interests to rely on.
But now consider this jolly little variation. (I should warn you at once that I can think of no substantive applications, but it's food for thought in any case.)
Suppose that Sender's type is still drawn from U[0,1], but now we think of it as being defined on a unit circle, with distance defined as the shortest possible way round the circle. So now, for example, choosing a = 1 when t=0 would be a very good action for Receiver, as this would give a-t=0.
Now there is no equilibrium in which Senders with t in different sub-intervals send different messages, no matter how small c is. (I am not sure I can prove that there is no informative equilibrium at all. Perhaps there's one in which a continuum of possible messages are sent, informing R about the value of t modulo c, or something similar.)
Proof
Suppose that there are n different sub-intervals around the circle; Senders with t in any of these sub-intervals send different messages. Let's take an arbitrary pair of adjacent sub-intervals (x,y) and (y,z), sending messages m1 and m2. As usual, since t is uniformly distributed, Receiver's best responses are a1 = (x+y)/2 and a2=(y+z)/2. And, as usual, because Sender's utility function is continuous w.r.t. a, Senders at the boundary where t = y must be indifferent between a1 and a2.
Therefore, it must be the case that the distance from a1 to y+c (Sender's ideal point when t=y) is equal to the distance from a2 to y+c. And so the boundary point y must be closer to the midpoint of (x,y) than to the midpoint of (y,z); and so the interval (y,z) must be larger than (x,y).
But this must be true for all intervals round the circle, which is clearly impossible. (If there is a finite number of intervals, n, then the nth interval must be bigger than the n-1th interval, which itself is bigger than the n-2th interval and so on down to the first; but it must also be smaller than the first interval which is adjacent to it. If there are an infinite number of intervals, then choose an arbitrary selection of 2 or more points on the circle and consider that the size of the intervals containing these points must continue to increase as you move clockwise around the circle. I think this works. Hey, it's a blog, and I'm only writing this to avoid real work.) QED.
Note that nothing here depends on the exact geometry of the "circle". The proof works for any interval with distance defined in an appropriately circular fashion. (But, as I said, what the hell are the substantive applications? A hunter is chasing an elephant, which itself is chasing a bear around a big lake; he wants his friend to slay both elephant and bear, but his friend only wants to tackle the elephant. Oh Lord. Answers on a postcard?)
and on the other hand
Sunday, 28 August 2005
It's 2 am. On the world service: "Reporting Religion".
First up, two rabbis discuss the meaning of Zionism. One of them believes that compromise over the status of Jerusalem is impossible, because the Jewish people have a contract with God.
Next: Delhi is being overrun by monkeys, who have been e.g. attacking children. The monkeys cannot be harmed because they are sacred, in fact divine.
And finally, does the Koran contain truths only recently discovered by Western science? A columnist in an Egyptian newspaper thinks so, and has a page of scriptural reinterpretation every week.
The traditional response to idiocies of this kind has been something like "religion and science deal with different, incommensurable kinds of truth". Here's an alternative way to look at it. All the three ideas above involve empirical claims, which are more or less testable... and completely preposterous. Monkey Gods, contracts between God and his "chosen people", quantum physics hidden in the Koran code: what serious person can regard these ideas with anything but contempt?
There are some complex explanations of the "revenge of God", the resurgence of religion over the past fifty years. Here's a simple one, which I'm sure cannot be the whole truth: enlightened, thoughtful people took their eye off the ball. The others got so angry when we disputed their stories, that it became easier just to go along and to pretend that there was no fact of the matter in these areas. So the nutters grew louder, more confident and madder than ever. Now we should learn our lesson. When someone talks nonsense, we should stand up and call it nonsense to its face.
Monday, 22 August 2005
In NY again
Friday, 19 August 2005
Quote
-- John Stuart Mill, The Subjection of Women (1869)
Monday, 8 August 2005
A Room of One's Own
I've just read Virginia Woolf's A Room of One's Own for the first time. What an astonishing book. It's only about a hundred pages long, but like one of those pop-up birthday cards, it unfolds to reveal a very intricate architecture, containing in miniature almost the whole programme of feminism from that day to this: anger at inequalities of power, wealth and privilege, and examination of their roots; a call for a women's history; a sceptical focus on men's “knowledge” of women; mistrust of simple ideas about the sexes; interest in relations among women where men stop being the focus. As if she just knew! The book – it was originally meant for a lecture to undergraduate's at the women's colleges of Girton and Newnham – is full of casual asides as to how a student might profitably look at, say, the domestic history of the Elizabethan household, or the psychology of women's art.
The idea that culture has material foundations, that you need a room of your own and five hundred pounds a year to be a poet, is a Bloomsbury theme, it comes out in Howards' End too. The Bloomsbury people's intellectual elitism and passionate love for higher things combined with this intense awareness to make them more egalitarian, funnily enough. If you think pushpin is just as good as poetry then the fact that some people are poor and others rich doesn't really matter much, especially as everybody is getting richer together. But if you think Art and Beauty and above all good Conversation are the most important things in the world, then it matters terribly. (So I suppose that the Blair government's artistic policy of “access” is another way in which the powerless little clique has conquered the world.)
Did they conquer the world? If you could trace back the lineage of the spiky-haired gothic brats in Camden Town, and the elegant loft-dwellers of Manhattan or Hoxton, and the pot-smoking first-homers in small towns everywhere – of everything alternative or bohemian or unconventional, which by now includes almost everything in contemporary culture, would you find the little Cambridge gang of lesbians and nutters? I would like to think so.
Here are two quotes which make me very happy.
Meanwhile the wineglasses had flushed yellow and flushed crimson; had been emptied; had been filled. And thus by degrees was lit, halfway down the spine, which is the seat of the soul, not that hard little electric light which we call brilliance, as it pops in and out upon our lips, but the more profound, subtle and subterranean glow which is the rich yellow flame of rational intercourse. No need to hurry. No need to sparkle. No need to be anyone but oneself.
So long as you write what you wish to write, that is all that matters; and whether it matters for ages or only for hours, nobody can say. But to sacrifice a hair of the head of your vision, a shade of its colour, in deference to some Headmaster with a silver pot in his hand or to some professor with a measuring-rod up his sleeve, is the most abject treachery, and the sacrifice of wealth and chastity which used to be said the greatest of human disasters, a mere flea-bite in comparison.
Reading books like this could be a very bad idea. After spending a few hours with Virginia Woolf, you may not want to spend time with anyone else, and you may not care about anyone else's good opinion. Then you will be cursed with the worm that dyeth not, of permanent dissatisfaction, because there aren't many people like that in the world.
But she's wrong about the silver pot. The trick is to want praise and commendation from the right people.
Sunday, 7 August 2005
Saturday, 6 August 2005
Thursday, 4 August 2005
You know you're at ICPSR when...
Good night!
Friday, 29 July 2005
Monday, 25 July 2005
i can't stand this tropical heat no more tracy!
a (slightly altered) political sticker in Ann Arbor. This is a very liberal town, not like the flyover stereotype.
it is HOT. and HUMID. I am drinking gallons of water, unable to concentrate or do anything.
things I didn't mention. When we were out dancing, two beautiful cornfed teenagers randomly came up and started shaking their booties at us in the most shocking way, bending right over a la detroit kronk. (If that's the expression?) They were chortling throughout.
So like previous generations of Englishmen, I am impressed with the frank and natural manner of American ladies.
What else? Well, the architecture isn't up to much and I haven't yet found a good breakfast spot. But thank god I have found a place to stay with aircon.
Sunday, 24 July 2005
We go out. A band playing greatest hits of the 70s - funk, country and soul indiscriminately. This is unique. I have never been in a place where I was the best dancer. People standing around and watching. Couples where the girl wants to get into it and the guy is just standing there, ass rooted to the floor. She thinks she's a free spirit being held back, but she's playing a role too and is quite comfortable in her self-imposed restraint! So we show them what it's all about.
.... and then we go outside and there's a black guy playing his guitar, a two chord blues version of the latest dance hit, I think, and some couples doing the best, most joyous dancing I have ever seen. They jive to his blues - Asian, Latino, Black and White, the crowd clapping - just out on the sidewalk! Unimaginable in Europe.
But then you realise they are all professional.
But then, here anybody can be professional. No certificate required:all you have to be is good.
You see, this ambivalence is not really going to go away. It can be so amazing and then so dull.
Sunday, 17 July 2005
Harrogate
I'm off to New York, then Ann Arbor on Thursday, where I'll be attending the University of Michigan Summer School - course on advanced game theory. Perhaps I'll finally develop some half-decent formal modelling abilities.... Stranger things have happened. There are many cool things at UMich, as well as Robert Axelrod (the Evolution of Cooperation guy) and Leda Cosmides and John Tooby (progenitors of evolutionary psychology) they have Juan Cole, a historian with a great blog about the Iraq war. (Hmm, scribbling this on Chris' PowerBook which although very beautiful does not seem to have a web browser that lets me put links into blogger. Juan Cole is at http://www.juancole.com .)
Wednesday, 13 July 2005
SPSS is a crock of crap
Tuesday, 12 July 2005
(The context is a discussion of cues and heuristics in voting - ie how voters can make decisions without being hugely knowledgeable about politics. But I think this is interesting anyway. It is partly inspired by the economics literature on "informational cascades". This is just a very simple example.)
If I trust that you know better than me how to vote, I may vote correctly. But I am certainly not helping make the democratic majority decision any more accurate. Under certain conditions, I may even make it less accurate.
Here's a simple formal example. Suppose that there are three voters on a simple Yes-No issue. One of these outcomes is unequivocally the right one. The voters each receive some information (a signal) about whether Yes or No is the right outcome. The accuracy of those signals varies: the voters have probabilities 0.7, 0.75 and 0.8 respectively of getting accurate information which suggests they should vote in the correct way. Also, these probabilities are known to all the voters in common, although the signals themselves are private. Suppose that the voters vote independently, and try to vote the right way. They all then follow their signals, which are more likely to be right than wrong. The chance of a correct decision is the chance of all three voting right, plus the probability of three different majorities of two voting correctly: 0.7*0.75*0.8 + 0.3*0.75*0.8 + 0.7*0.25*0.8 + 0.7*0.75*0.2 = 0.845.
Now suppose that the best informed voter, Mrs Point Eight, announces what her signal is (which way her information points) before the vote. The other two voters are then each in one of two situations. Either their signal agrees with Mrs Point Eight: if so they vote the same way anyway. Or their signal disagrees. If so, they now have two conflicting signals. But Mrs Point Eight's signal is more accurate than theirs, so they sensibly choose to ignore their own signal. In either case, they vote with Mrs Point Eight. Unfortunately, this means that the chance of reaching a correct decision has gone down to 0.8.1 By trusting a better-informed person, the individual voters have become more accurate (their chance of being right is 0.8 instead of 0.7 or 0.6) but have made the collective outcome less accurate.
Saturday, 9 July 2005
Monday, 4 July 2005
Cambridge at the weekend
Anyway, it was a psychedelic evening. Thanks to Juliet Banfield for putting it on, and a big shout out to old friends and also new, or rather recently remade - hello Holly, Mike, Susie, et al. A night when I make a spectacular ass of myself, my dears, can still give me such intense happiness that I spend most of the drive back home actually singing. (Perhaps this is bad in some way?)
As usual, there are photographs, some of which involve Dave Padua and are sexually ambiguous. Watch this space.
Friday, 1 July 2005
Tuesday, 28 June 2005
If you have the Firefox web browser, and regularly use Google Scholar, then check out this wonderful little bit of tech. It basically adds a "saved searches" box to Google scholar, which lets you see if new articles have been added. Very useful for monitoring research. You'll need the Greasemonkey extension for Firefox. I love this idea of consumers customizing web pages.
Then maybe you should read this red hot little number. (Just something I'm looking at at the moment. No, really, you do not have any reason to read this document.)
Thursday, 9 June 2005
Game theory of killer initiatives and legislative responses to initiatives
(Just something I jotted down - a sample of the kind of work I do.)
A common institution in the context of direct legislation is to allow the legislature to respond to a proposed initiative. I put forward a game-theoretic rationale for having this institution, but point out some potential problems. The problems also apply to cases where interest groups outside the legislature sponsor “killer initiatives” which are drafted so as to affect other initiatives on the ballot box.
The rationale is as follows. We consider policy on a single dimension. The status quo is SQ and voter median ideal point is at MV. All utility functions are given by distance:
utility = |outcome – ideal point|
We assume that proposing an initiative costs C, and that the potential proposer is extreme, with ideal point marked PIP. MV' marks the point [MV + (MV -SQ)], i.e. the median voter is indifferent between MV' and SQ. To simplify, we assume that indifferent voters vote “yes” on all proposals.
SQ--------------MV-------------MV'----------------PIP--------
If the legislature may not respond to the initiative, and if MV' – SQ > C, then the proposer proposes an initiative at MV', which the voters accept. Policy is now as far from the median voter as it was before, but in the opposite direction.
Now suppose that the legislature can respond to proposed initiatives by bringing forward its own counter-proposal. If it does so, then the counter-proposal, rather than the status quo, will be the result if the ballot proposal fails. The legislature incurs no cost in bringing forward a counter-proposal: this seems reasonable, as it has no signature requirement to fulfil. Most likely the legislature's ideal point LIP will be at SQ: however, we consider all alternatives.
SQ--------LIP-----MV-------------Mv'----------------PIP--------
If LIP < MV: any proposal to the right of MV can be countered by a proposal to the left of MV, but closer to MV. Therefore, if MV – SQ > C, the proposer proposes MV; otherwise the proposer does nothing.
If MV' > LIP > MV, any proposal to the right of LIP will be met by a counter-proposal at LIP. Therefore, if LIP – SQ > C, the proposer proposes LIP; otherwise the proposer does nothing.
If PIP > LIP > MV' (ie legislature and proposer are both extreme), and if LIP – SQ > C, again the proposer proposes LIP or any proposal to the right. The legislature then responds with LIP and offers the public the unappealing choice between two proposals which are both less preferred to the median voter than the status quo. The result is LIP.
If LIP > PIP > MV' then proposer proposes PIP and legislature responds with PIP or any proposal to the right; here again, proposer and legislature collude to offer the public a Hobson's choice. The result is PIP.
In the latter two cases, a rule like the one outlined makes outcomes worse from the point of view of the median voter and the associated majority in his or her lea. However, these seem rather implausible: why shouldn't the legislature legislate directly, rather than take the roundabout initiative route? In the more plausible first two cases, the result is always better for the median voter. Allowing legislative counter-proposals essentially turns the contest into a classic Downsian two-horse race.
Potential problems
The institution as proposed above is rather simple. However, it doesn't always work like that. Switzerland allows the legislature to bring forward proposals in response to an initiative, but rather than replacing the status quo, they are voted on separately. If both the original initiative and the legislature's response pass, the legislature's proposal overrides the original initiative. (I think. Hey, this is a blog entry.) In a less institutionalised, but broadly similar way, “killer initiatives” are sometimes seen in the US. These are drafted so as to override the effect of another proposed initiative: again, if both pass, the killer initiative will specify limitations to the effects of its target.
We can model both these situations similarly, as proposers 1 and 2 decide in turn to propose initiatives. If both proposers introduce initiatives which pass, we assume that 2's proposal is implemented, as it contains clauses designed to nullify the effect of 1's proposal.
The choice situation facing the voter now becomes complicated and may require sophisticated voting – something that, it is often believed, voters in a direct democracy find difficult (e.g. Lacy and Niou 2000). That paper points out a similar result when preferences over two initiatives are non-separable. In this case, preferences over outcomes are single-peaked and the space of outcomes is one-dimensional: the problems arise because voters don't know what the effect of their vote will be. As an example, consider the following choice situation where two initiatives have been proposed:
---SQ--------------P2-------------P1---
Given single-peakedness, there are four possible preference profiles for voters:
SQ > P2 > P1
P1 > P2 > SQ
P2 > SQ > P1
P2 > P1 > SQ
The last two groups of voters have no problem: they can vote yes on P2, and yes or no on P1. A voter who happened to be decisive on either outcome could not possibly experience regret, as they would always have brought about an outcome which they preferred to the alternative – either by achieving the best possible result, P2, or by preventing P1 from supplanting the status quo, or by possibly allowing P1 to win (assuming P2 was not implemented).
The “extremists” in the first two groups are not so lucky. Consider first those who wish to preserve the status quo. A voter (or group of voters) who happens to be decisive on P2 faces the following set of choices:
P1 outcome | Vote | Result |
Passes | Yes | P1 and P2 pass: outcome P2 |
| No | P1 passes: outcome P1 |
Fails | Yes | P2 passes: outcome P2 |
| No | neither pass: outcome SQ |
Clearly, if P1 is going to fail, it's better to vote no on P2 and preserve the status quo. Equally, if P1 will pass, you should vote for P2 in order to salvage something.
Partisans of P1 face a similar problem. They should vote for P2 if P1 is going to fail, but against P2 if P1 is going to pass. The problem in terms of votes is analogous to that mentioned in Lacy and Niou (2000). For example, a member of the first preference group has the following preference ordering for outcomes of the votes:
P1 | P2 | Result | Ranking |
N | N | SQ | 1 |
N | Y | P2 | 2= |
Y | Y | P2 | 2= |
Y | N | P1 | 3 |
Thus, despite single-peakedness over the actual issue, we end up with non-separable preferences over the outcomes of votes.
Wednesday, 8 June 2005
These are for myself but also public.
R project for statistical computing, in case you didn't know.
Adding a straight regression line to a plot
model = lm(y ~ x)
plot(y ~ x)
abline(model)
Adding a wiggly line to a plot
plot(foo ~ bar)
lines(lowess(bar, foo))
Tuesday, 7 June 2005
Friday, 3 June 2005
New York
The Guggenheim staff told me off for this.
For a few months I was a volunteer in the Simon Community, which deals with rough sleepers in London. There was a resident there who was a lot of trouble. He was very shrewd and insightful, and he could be extremely aggressive. Anyway, if he thought he was being messed around somehow , he would look straight at you and say: "I'm not a prick."
Brussels tried to get the European electorate to vote for a long, turgid, incomprehensible constitution - oops, sorry, constitutional treaty. (It was a constitution because everyone had to vote for it, but it was long and boring because it was a treaty. Got that?) If they didn't, the sky would fall in, there was no plan B, et cetera. To which the reply came: "I'm not a prick."
One more thing while I'm ranting. Various important Brussels people have been publicly puzzling, agonizing, over what the two No's can possibly "mean". Here's a clue: they mean No, you numbwits.