Ryan Giordano in London, UK

Taking EC484 for the LSE EME? Here Are Some Notes.

noreply@blogger.com (thegio) — Mon, 19 Jan 2009 19:07:00 +0000

Before throwing away my notebooks and before I forget everything, I wanted to write down some notes on what I learned throughout the year about EC484, the core econometrics course of the LSE EME MSc. Not that any of my friends who read this blog will need them, but I'm being indexed, and maybe I'll come up on some search results for someone someday.

If you're taking the class and want to check these out, you should take them about as seriously as you would take advice from any of your other classmates, which is not all that seriously. I did write them up in a hurry and over six months after taking the exam. Still, these are the notes that I wish I had had in January, and I honestly think they will help most people taking the course in at least some small way.

Note that they are essentially useless without Professor Robinson's course packs, to which they are essentially just commentary and supplements.

Here they are:
EC484-Giordano Supplemental Notes.pdf

Notebook Pages.

noreply@blogger.com (thegio) — Mon, 19 Jan 2009 13:15:00 +0000

It's hard not to have an emotional attachment to the piles of old notebooks you accumulate in grad school. The handwriting, the mistakes, the sketches, and the order of the notes all tell a long, hazy story legible only to the writer. However, airline luggage weight limits force you to let that attachment go. To soften the loss, I'd like to post some typical pages from the last two years. After posting this, I'll go throw them all away.

This is a very typical page. I've always thought that math looks pretty when viewed from a short distance away so that the ideas are illegible but the orderliness is not.

For some reason I developed a special flowery handwriting for all my notes in EC475, my second-year quantitative economics course.

I have lots of sketches of the backs of peoples' heads.

Taking Chinese gave the notes for some of the slower-paced classes an extra flourish. I seem to have spent a lot of time practicing Chinese in graph theory.

By looking at the subjects I was working on and recalling my exam schedule, I can find the first two pages I wrote in my notebook after missing my game theory exam, and they make me proud. Here they are: the core reading list and first problem set from ST422, the time series exam that I thought I might be able study for a week and take instead of the game theory one. In the notebook, there was no fussing and no nonsense: it went straight to what needed to be done. Seeing this makes me feel proud.

Some Recommendations.

noreply@blogger.com (thegio) — Wed, 07 Jan 2009 16:49:00 +0000

I have been meaning for a while to write down some recommendations before leaving England, so here they are.

Regent's Canal East of Camden Town
Starting at Camden Market, you can walk, in a single afternoon, all the way East along the canal until you get to the river, and in the doing see an incredible range of London living, from posh oligarch playhouses to grimy street art with plenty of peculiar pubs to stop at along the way.

Living Lounge Hostel (Lisbon)
If you must leave London, consider going to Lisbon, Portugal, to stay in the best hostel ever of all time, the Living Lounge Hostel. (The same people operate a similar hostel down the street called the Lisbon Lounge, which is why the website has that name.) Idiosyncratic rooms, an artsy lounge, delicious in-house dinner, and jovial hipster staff made us wonder why we were not being charged high-end hotel prices to stay here.

Jeremy Bentham
Don't look it up and don't ask any detailed questions. Just go to the University College of London campus by Russel Square and ask to see him. You will be told where to go.

Monmouth Coffee
If you need good high-quality American style coffee in London, there is only one place to go (well, three, since they have three branches): Monmouth coffee. For two years I went out of my way to buy all my coffee here. They'll even sell you raw beans if you want to tempt the fire alarms in your dorm with a home-roasing project.

Sir John Soane Museum
The walk from the entrance hall to the collection chambers of this free museum near the LSE campus is my idea of what the onset of madness must be like.

Cotswolds Bike Trips
My favorite thing of all to do in England is cycling in the countryside, and there's no better place to do it than the Cotswolds. You can get a train with your bikes from London to the West side (e.g. Cheltenham Spa), cycle to a remote bed and breakfast in the middle using a trusty cycling map, and catch the train back from the other side (e.g. Moreton-on-Marsh). Be forewarned: Google maps is worthless for this kind of thing.

Jamón Ibérico
The best food. I suspect the Spanish may have invented it to piss off the Moors. Occupied by Muslims? What better way to get back at them than developing the most delicious pork the world has ever known?

Nine Elms Market
There's a sort of hidden Sunday market behind the Vauxhall Sainsbury's. It goes by the same name (Nine Elms Market or New Covent Garden Market) as a wholesale early-morning market, but it operates on Sundays until mid-afternoon and doesn't seem to be advertised anywhere. I believe this is the cheapest place to get anything in London. from housewares to giant wheels of cheese to bike parts to giant slabs of bacon to bowls to Dancehall CDs to sugar cane. My friends once bought half a goat there. There's great street food, too. To find it, just follow the crowds disappearing through the gate in the fence behind the Sainsbury's parking lot.

My Blog: #2 For the Flag

noreply@blogger.com (thegio) — Wed, 10 Dec 2008 22:49:00 +0000

My mom and I share a web hosting service, and the bills get delivered to her because she has a fixed address (and I have no money). Recently, she sent me an email to tell me that we had been charged for excess traffic on our shared account. Since our account comes with 15GB per month included, that sounded absurd, and I promised to look into it. A couple weeks later when I actually did, I found that my blog had, indeed, generated over 17GB of traffic in October, almost all of it being people downloading the little American and Kazakhstani flag pictures I have on my site.

Sure enough, if you search for "American Flag" or "Kazakh Flag" on Google images, my website is number two for each of them. (Embarrassingly, both seem to be jpeg, too. Now what did I do that for?) I suppose I got "lucky" and at some early point and had my pictures show up high on the list of image results, and people are usually as happy with one flag picture as with another, so they got clicked on a lot, which reinforced Google's notion of their appropriateness, and now I am serving the two nations flags to the world.

Up till the point where we got this extra bill, the only effect of this was that I would sometimes get emails from junior high students who were doing some sort of internet project and had been told by their teachers that they had to receive permission to use any digital content downloaded from the internet. They would ask me if they could use my flag picture and I would make fun of them (or their teachers), and that was all fine.

But now that I have to pay for that bandwidth (well, for now, my mom is paying, actually, but that will change soon I think) it's less entertaining. And for a while I've been thinking I should sign up for AdSense, Google's website advertising service, to get an idea of what it's like in advance of working there. This provides a perfect excuse to do just that. So as of today, when you visit my old Peace Corps blog you will see ads on the left hand side. Right now it seems to be serving ads for flights to Kazakhstan, which is just fine with me.

I have never wanted to profit financially from my Peace Corps experience, so after defraying my mom's bandwidth bill, I will donate all the proceeds to volunteer small project grants. How am I generating the economic value behind these proceeds? Only through the Miracle of the Internet, about which I am sure to be thinking a lot in the months ahead.

Watch My Hands, Find the Asset.

noreply@blogger.com (thegio) — Sun, 16 Nov 2008 14:04:00 +0000

Recently I spent some time learning about the ABX-HE index, a financial index that tracks the performance of mortgage-backed securities. It is designed, among other things, to improve "transparency" in the market of mortgage backed-securities.

So how is this transparency-improving index calculated? Although the legal contract is available on the website of Markit, the company that calculates and compiles the index, they don't seem to provide a clear and concise description on their website of how this is actually done. So with that caveat, as far as I can gather the index works like this:

- Take a lot of home loans to subprime borrowers
- Package them into bundles, each bundle with its own definition of what a loss on the bundle of mortgages means
- Select twenty such bundles through a voting process by index participants
- Divide each bundle into tranches (with one tranche bearing the first, say, 5% of the losses, the next tranche bearing the next 10%, etc.)
- Take credit default swaps (CDS) on the tranches (that is, a par value contractual obligation to pay in the event of a loss in exchange for a fixed coupon payment)
- Bundle these credit default swaps together to form tranches of the index
- Take the remaining principle weighted average payments on these credit default swap contracts as a floating leg and pay a fixed leg based on the coupon payments at the start of the index
- The index will be 100 minus the amount you would pay to acquire the floating leg of such an average.

So the index is a price-based index of a basket of CDS options on tranches of a bundle of contractually defined losses on subprime mortgages. A long way from the ol' Dow Jones, aren't we?

I Am My Own Little Hoodrat Friend.

noreply@blogger.com (thegio) — Fri, 14 Nov 2008 20:26:00 +0000

It has recently come to my attention that the building I live in consists predominantly of council flats, which is to say, in British parlance, that I live in the Projects. Fortuitously, my friend Matt Zanon has recently sent me a parcel containing, among other things, a copy of N.W.A.'s seminal LP "Straight Outta Compton" featuring Dr. Dre's, Easy E's, Ice Cube's &ct's reminiscences of locales such as mine. Straight out of Compton, are you Doctor? Well, I am straight out of Stockwell.

I have recently made a point when walking around my "hood" (neighborhood) to "bang" (play on my ipod) this album, in case I should have to quickly indicate to any of my "homies" (neighbors) that I too am "ghetto" (live here). To further "drive home" (demonstrate) this point, tonight Sofia, Gertje, Nina, and I will be attending the local Roller Disco, which I understand to be our nexus of "Urban Culture" (city culture). Perhaps while I'm there they will "bang" something by Ghostface Killa and I will grin a knowing grin, for I too am now from the "Projects".

What Is Next.

noreply@blogger.com (thegio) — Sat, 01 Nov 2008 16:36:00 +0000

Of all the jobs that might have been available after finishing LSE, I've been offered one of the best imaginable -- statistician at Google. My future coworkers are great, the work will be a combination of research and practicalities (all undertaken with one of the largest data sets in the history of mankind), and my salary will be at a grown-up level for the first time in four years. I'll be moving to Mountain View the first week of February.

Although I couldn't have hoped for a better job offer, I'm very sorry to be leaving London, which is the coolest place I've ever lived. And of course, it's going to be very hard living away from Sofia. I've asked to delay my start until February in order to have a couple months with her here (her visa runs out at the end of January), and in the mean time I'm working at Macquarie on a temporary contract. With luck, she'll at least be able to come help me move, and I can take a vacation to Buenos Aires next year.

So pretty soon, I'll be putting those old flowers in my hair and heading out to San Fransisco...

What Is Going On.

noreply@blogger.com (thegio) — Fri, 11 Jul 2008 21:03:00 +0000

On the very unlikely chance that you read this but don't communicate directly with me, here is what is going on.

The bad news first: due to my own stupid scheduling error, I missed an exam. In the US, this would mean a resit a week later and possibly a small penalty in the exam grade. In the UK, it means you fail all your courses by default. Fortunately, the school has taken pity on me and will allow me to resit -- in a year's time -- without further censure. As a consequence, I will not officially graduate until next year.

The good news is that all my other marks were distinctions, meaning after I pass my remaining half-credit in 2009, I will have finished the econometrics MSc at the LSE with top marks.

In the meantime, I am working at Macquarie in London as a summer intern, learning about banking and thinking about what to do next.

A New Business.

noreply@blogger.com (thegio) — Fri, 11 Jul 2008 21:02:00 +0000

Sofia tells me that the Argentinian proverb goes "A bird in hand is worth a thousand in the sky".

I think it's time to start a bird-in-hand export business.

An Econometrician Joke.

noreply@blogger.com (thegio) — Sun, 08 Jun 2008 12:51:00 +0000

Thanks to Abby Samp:

"Three econometricians went out hunting, and came across a large deer. The first econometrician fired, but missed, by a meter to the left. The second econometrician fired, but also missed, by a meter to the right. The third econometrician didn't fire, but shouted in triumph, 'We got it! We got it!'"

My Paper Is Done.

noreply@blogger.com (thegio) — Wed, 30 Apr 2008 21:52:00 +0000

LSE Econometrics students don't write a thesis. Instead they write an "extended essay" that counts for half of the grade for one of their four courses. I submitted my paper on Monday with the weighty title of "Matching CEX Consumption Moments With Bayesian Learning of Heterogeneous Income Profiles". Read it if you dare.

I have mixed feelings about it. On the plus side, it was the first time I brought a research idea that was all my own from conception to fruition. I think it was ambitious for a six week project, and I managed to get results with my own code and data. I did go beyond simply replicating someone else's results, which can be difficult to do at the Master's level. I think overcame significant obstacles on the computation side. (Which is to say that managing to do anything computationally intensive with LSE resources is an accomplishment in itself.) And, finally, this is my first ever economics course, so maybe the world will be lenient with me.

On the negative side, this is exactly the kind of economics research that I dislike. I am essentially studying a mathematical model rather than the real world. The assumptions of the model I chose are excessively removed from reality (it is a very similar model to the one I complained about in an earlier post, "Do You Make Your Consumption Decisions This Way?"). And my conclusions are particular to this model in this configuration, without any clear way of relating it to similar problems.

I guess the lesson is that it's easy to complain about the way economics research is done but hard to do better, at least not the first time you try, anyway.

Now that I'm going to start working after graduation, I won't have time to take this further. But if anyone else is inclined, I have available a very nice, modular, thoroughly commented matlab code base to do all the stuff in this paper...

Макинтошпен қақақ әріптерін жазу мүмкін!

noreply@blogger.com (thegio) — Thu, 24 Apr 2008 08:42:00 +0000

At last! Multilingual Mac provides a Kazakh language keyboard for the Macintosh!

Can You Solve These Two Chess Problems in Five Years?

noreply@blogger.com (thegio) — Wed, 23 Apr 2008 12:34:00 +0000

While living in Boise, after I first started playing chess with Mike, I bought a copy of this book, Winning Chess Tactics for Juniors from the used book store in Hyde Park. It is nothing but hundreds of three-move chess puzzles to help practice different tactical techniques. I have always found them difficult, but after many years of keeping it in the bathroom (and so trying a few every day), I've made some slow progress, which means I can now actually solve about half of them.

On the cover of the book there are two problems, with the challenge, "Can you solve these two chess problems in thirty seconds?" When I first got the book, I stared at them for days and never saw anything. These two puzzles alone in the book don't have the solution included, so I'd always found them frustrating. After a long while, I finally took another look at them last week, and found that, with a lot of effort, I could solve both of them. From the moment I first saw the problems it took me about five years, approximately 5.2 million times longer than thirty seconds, which is a fair measure of my innate ability for chess.

The Best Fruit.

noreply@blogger.com (thegio) — Thu, 17 Apr 2008 19:05:00 +0000

The votes are in (that is, my and Sofia's votes). And the Best Fruit is unanimously: The Mango. There is no other fruit better than The Mango.

4767 Message, 4765 Unread

noreply@blogger.com (thegio) — Tue, 01 Apr 2008 10:41:00 +0000

I have an email account associated with my webpage, thegio at thegio dot net. It was this address that I gave all my friends from Kazakhstan before I left. Every now and then, my counterpart sends me an email at my gmail account, and she usually complains that I don't write back to my students when they email me. This isn't true -- on the few occasions that I do hear back from my old students, I write them back right away. I always chalked this up to meaningless motherly scolding on the part of my counterpart.

But a little while ago, I noticed that if I sent myself an email to the thegio address, it didn't get forwarded, although I did receive some things from the address, such as gmail news updates. Today I found the problem -- the address has been massively spammed, though I didn't realize it because of gmail's spam filtering. However, short, badly spelled messages from obscure servers -- precisely the sorts of messages I would get from my students -- didn't make the cut. A brief inspection found two messages that from former students that I didn't receive in my gmail account.

So now I basically am going to go through all 4.7 thousand messages and separate the wheat from the viagra by hand. I don't feel that bad about it, though. At least now I know what has been going on and can correct it, and perhaps be in closer touch with my Kazakhstani friends.

I Did Not Like Professor Quah's Presentation.

noreply@blogger.com (thegio) — Tue, 04 Mar 2008 10:31:00 +0000

Our economics department head, Danny Quah, is a clear and engaging speaker. I had seen him give an interesting, if a little fluffy, presentation at the LSE series "Thinking Like A Social Scientist" (which series, Prof. Quah's talk aside, has done more than anything else to convince me that I do not want to think like a social scientist), so I was looking forward to hearing him give a more substantive presentation for an economics-only audience. Yesterday I got the chance. Even better, his talk was about development and growth, which are very interesting topics to me.

The presentation went as follows. As a preface, he said that he preferred to take a data-oriented approach to studying growth rather than a model-oriented one. Then he began with some basic statistics about the phenomenal economic growth and inequality growth in the last one hundred and fifty years to motivate his talk. He then presented the Kuznets curve, an empirical curve that suggests that countries that have low inequality and low levels of development proceed to higher levels of inequality as they develop and industrialize, but at some point turn around and become more equal as they continue to develop. He then presented an economic model that could explain this wherein non-convexities in the production function lead to endogenous inequality. (This was essentially a mathematical form of the poverty trap hypothesis.) After this, he turned to the effects of trade on inequality, presenting a model in which a high-technology society and a low-technology society begin to trade with one another, resulting in greater inequality in the latter and less inequality in the former. Finally, he showed graphs of income inequality in the US and China that seemed to debunk this model, since they both grow over time as globalization proceeds.

Unfortunately, for a talk that was prefaced with praises of empiricism, the models Prof. Quah discussed were only compared with only the coarsest stylized facts. For the first model, we had only the Kuznets curve -- that is, that countries seem to become more unequal as they develop. To begin with, the Kuznets curve is based mostly on single-time multi-country data, not time-series data for a single country, so it's very unlikely that it describes an economic process that a single country proceeds through. But even if you take it at face value, presenting a single stylized model that can describe the generation of inequality does not inform your understanding at all without some effort to distinguish the evidence for this particular model from all the other potential explanations of the curve. Similarly, presenting a two-country world economy with factor endowments does not really inform our understanding of the world unless the evidence of its truth is discussed at length. At least Prof. Quah presented some very coarse evidence against this model. However, the model was so simple that you can argue away the difference from reality by saying that important particular considerations overwhelm the model's simple predictions (as he did for the case of Central America). Given this, and its non-conformance with US-China inequality data, why study the model in the first place?

My problem with this talk, which is a problem I have with almost every academic economics talk I attend, is that learning about a model is not the same as learning about the real world. It's easy to forget this when you spend an hour carefully trying to follow the algebra of an articulate and widely respected professor. But at the end of the day, talks like this, in my opinion, provide almost no added understanding of the real world. Economics is so complicated, and its data so rich and often counterintuitive, that discussion of it should be ten percent theory and ninety percent data. Instead, in academics, it seems to be almost always the other way around. If this is what it means to think like a social scientist, I think I'll stay a scientist.

Now, I'm sure Prof. Quah doesn't do his research like this. Doubtless he is a very clear empirical thinker and has simplified things for an audience of students. Also, he has another follow-on presentation to make on Wednesday, and perhaps then he will tie this first talk in with reality a little more tightly. Nevertheless, if this is how budding economists are taught, this is how they will think when they go into the real world to practice economics. And I think history has ample enough evidence of the harm that can do.

Bayes Pays Off.

noreply@blogger.com (thegio) — Fri, 08 Feb 2008 20:24:00 +0000

In our stats class today, to make it fun, our professor actually did the experiment described in the last post, "Bayes Breaks Down", and I was lucky enough to be the one chosen to make the decision. I opened the envelope and it contained £12 - a tenner and a two-pound coin. I reasoned that if the other contained twice as much, it would require at least two coins (since there are no £4 coins, and no bills smaller than £5), whereas if it contained £6, it could also be done with a single note and single coin. Since he would have chosen the amounts in the envelopes to be as indistinguishable as possibly before being opened, it's unlikely that he would have put two coins in one and not in the other. So I chose to keep the first envelope.

It happens I was right. Sometimes priors pay off.

Bayes Breaks Down.

noreply@blogger.com (thegio) — Fri, 08 Feb 2008 19:49:00 +0000

In my class called "Developments in Statistical Methods", I sometimes feel like Harry Potter secretly practicing military magic. We are studying Bayesian methods, which are spoken of only in a whisper by established LSE econometericians. "No Bayesians here," they hiss.

In a word, Bayesian statistics allows -- indeed, requires -- you to make explicit your prior beliefs about the parameter you're trying to estimate. While Frequentists believe in one true data generating parameter that you get closer to with more data, Bayesians have a range of beliefs about the parameters that is continually updated as you get more data. As the amount of data becomes infinite, the two methods are usually identical. With small samples, the results can differ markedly. In fact, questions with very small data sets, like a single observation, are only addressable by Bayesian methods. Frequentists throw up their hands, saying that the notion of "probability" has no meaning for a single observation.

Consider this question. I show you two apparently identical envelopes. I tell you that both have some money in them, and that one has twice as much money in it as the other, but you cannot tell which just by looking at them. You then get to choose one envelope and open it, and in it you see £12. I offer to let you keep the twelve pounds, or take whatever amount is in the other envelope. You want to maximize your expected take. Which envelope should you pick?

Here is where you should think about it if you want to before reading further.

First, notice that there is an obvious correct answer. If you have absolutely no information about how much I'm likely to have put in the envelopes, it should not make any difference which envelope you pick. Any probibalistic framework that does not give you this answer is fooling you somewhere and failing to do what probability theory is designed to do -- that is, describe and quantify your uncertainty.

But how do you characterize what you don't know in this case? It is natural to say that there is a 50% chance that the other envelope contains twice as much and a 50% chance that it contains half as much. But then the expected value of switching is 0.5*24+0.5*6=12+3=15. 15 is larger than 12, so you should take the other envelope, since the expected value is higher.

Naturally, a Frequentist approach is no use here. So let's look at it from a Bayesian point of view. Suppose you have a prior belief about the distribution of the amount in the smaller envelope which I will call p(x), and an independent prior assigning 50% probability to either envelope having the higher amount of money. Having observed the amount in one envelope to be y, the posterior probability, q, of the money in the other envelope is given by

q(y/2)=p(y/2)/(p(y)+p(y/2))
q(2y)=p(y)/(p(y)+p(y/2))
q is zero otherwise

(To see this, note that the other envelope must contain either y/2 or 2y. Then the numerator is proportional to the prior probability that that the smaller envelope contained the value implied by the second observation, and the denominator just normalizes so the probabilities sum to one.)

Notice that whether or not you choose to switch depends on your prior beliefs about the distribution of the money in the envelopes. So if you want the first observation to be uninformative, that should mean that no matter what y you observe,

q(y/2)=q(2y) =>
p(y/2)=p(y) =>
p is constant

This is the assumption we made implicitly when we did the above expectation calculation. But notice that a constant over the entire domain does not integrate to one -- that is, this prior is not a valid distribution function. In fact, you cannot have a valid prior that manages to capture the true uncertainty here, which is that you really have no information about the second envelope having seen the first.

I really like this paradox. It brings to light two deep (in my humble opinion) ideas: probability theory's main role is to characterize what you do not know, and the way you characterize it may affect your results in unexpected ways.

A Remarkable Coincidence.

noreply@blogger.com (thegio) — Sat, 02 Feb 2008 11:24:00 +0000

I hadn't read a novel since winter break, having been preoccupied with school, when a week ago I was studying at the University of London library. I usually just pick a desk in the stacks and work on problem sets. This time, when I looked up from my statistics after a couple hours, I noticed that I was next to the Japanese literature section. I've been looking to read "The Wind-up Bird Chronicles" for a long time, and no public library seems to have it, so I thought I would check. The didn't have it, but they had another Murakami book, "A Wild Sheep Chase", so I checked it out.

Like any good book, it dominated my life for a couple days. It's a strange story (natch) which ends with a psychological anticlimax, the hero having lost everything, unsure of the value of what he accomplished, and without a way of going forwards. It was one of the books where you're sorry it ends where it does, although to be expected from Murakami, who specializes in driftless disquiet.

Then, yesterday, I had to go to the Quad (LSE's filthy version of an undergraduate student union) to see the entertainment officer. There is a good used book shop there, where I usually stop to browse (and usually buy something). This time, they had a Murakami book I've never heard of before, called "Dance Dance Dance". It was only £3.50, so I picked it up.

It begins with the author having dreams about the Dolphin Hotel. That's funny -- there was a Dolphin Hotel that figured prominently in the last book. Murakami must have a preoccupation with this hotel. And it's in Sapporo, just like the last book too. Maybe it's autobiographical, I think, and both of these books are getting at something that actually happened to him in Hokkaido. A girl told him they had to stay there. And it was part museum, the halls stuffed with dusty...fleece? Ah ha! It's the continuation of the last book!

There's nothing on the jacket to suggest that it's a sequel to an earlier book, nor was there anything in the other to suggest it was continued. I just happened to pick up the two books, one after another, in two different places, in the correct order, within a week. A fair real-life probability problem; or perhaps a real life Murakami novel?

I'm Here.

noreply@blogger.com (thegio) — Sat, 02 Feb 2008 11:20:00 +0000

There were some comments asking why I'm not writing anything, and today an old friend wrote to me, concerned that something might have happened to me. So I suppose I should say: I'm ok. I'm just kind of preoccupied with school, much of which I think wouldn't be that interesting to almost anyone. (The ample proof of this is written on peoples' faces when I first tell them I'm studying econometrics.) But there really are always interesting things that happen, it's only a matter of me paying more active attention to them. So I'll try, for my own benefit as much as anything.

同志 Is Not What It Used To Be.

noreply@blogger.com (thegio) — Sun, 18 Nov 2007 22:31:00 +0000

I was studying Chinese in the kitchen when Guoming (国明! Really! She has a textbook name!), my Chinese flatmate, walked in. I wanted to practice, so I asked her in Chinese, "Are you an overseas Chinese? (华侨）?" She said she was not. "So you are a real Chinese person?" Yes, she said. "So may I call you comrade?" ("同志", literally "same will".) She said nobody says that anymore. "But is it ok if I call you that?" Here, she switched to English, and she said, "Do you know why I don't want you to call me comrade? It has a second meaning, and nowadays, it is used for homosexuals."

Do You Make Your Consumption Decisions This Way?

noreply@blogger.com (thegio) — Wed, 14 Nov 2007 23:02:00 +0000

We are studying consumer behavior in my quantitative economics course. The model we are working with assumes that

1) Consumers gain utility from consumption, and act to maximize that utility given a budget constraint
2) They receive labor income and have the opportunity to save and receive interest or, in some cases, to borrow from future expected income
3) They consider future utility worth less than present utility, and discount it at a constant rate over time.

These are not, by and large, loony assumptions (except for the constant future discounting rate, which is certainly loony, but let's ignore for now). Given all these parameters, it's easy to write down the equations that give the consumption and saving patterns that maximize your utility. These equations relate the amount you consume today to the expected amount you will consume tomorrow. I say expected, because you don't know certain things about the future: in particular, you may be uncertain about your future income (maybe you face a risk of getting fired, or maybe you will win the lottery, etc.).

If you think about this, you will see that it's a little problematic. I need to know how much I will consume today, and the optimizing equation says that I need to average out my utility tomorrow. My utility tomorrow depends on my random income, and also on my utility the day after that, which depends on -that- day's random income and the next day's utility, etc. on to infinity (or death). The mathematics get a bit hairy. And it's only tractable at all if you actually know the probability distribution of your future income stream and future income rates so that you can calculate the mathematical expectation.

Naturally, estimating these distributions and solving these models pays the bills for numerous economists. They are not easy problems. In fact, finding the optimal consumption behavior requires computer simulation, as it is not analytically tractable with common utility functions. (I spent today writing a program to do it, which is why all this is on my mind.)

What is crazy about all this is that I spent the day writing a computer program that is supposed to tell me how I, as an economic agent, am presumed to behave. If it requires the efforts of scores of economists and many years of simulation, mathematics, and statistical analysis to derive the behavioral consequences of this model, is it not absurd to think that people actually behave in anything remotely like this way?

Contrary to what some people seem to believe (these people are coming out of the woodwork to complain to me now that I am becoming an economist), economists are not stupid, and are fully aware of these shortcomings. However, it's hard to come up with a satisfying alternative description of people's behavior. If people don't rationally optimize the value of their consumption, what do they do? If they are not "rational", what governs their decisions? If they don't have complete information, what assumptions to they make about the probability distribution of future events? Even if you can tell stories about these things, it may not be trivial to make them into meaningful, helpful mathematical models. But the fact that these kinds of big problems are still out there is, to me, one of the very exciting things about economics right now.

Sandwiches Are Not a Crime.

noreply@blogger.com (thegio) — Wed, 31 Oct 2007 00:02:00 +0000

I would like a t-shirt that says "Sandwiches Are Not a Crime". Because it either suggests

(a) That there is a large class of people who believe that sandwiches are a crime and need to be dissuaded of this opinion or

(b) The idea that what is a crime and what is not should be decided by enumerating all things that are NOT crimes, the remainder being crimes, rather than the other way around,

both of which I think are interesting things for a t-shirt to evoke.

A Fascinating Mistake.

noreply@blogger.com (thegio) — Sun, 14 Oct 2007 19:04:00 +0000

I'm going to enjoy my quantitative economics course (EC475). Our first handout described the odd results of Henry J Moore in 1914 when he tried to measure the slope of the demand curve for pig iron. When he regressed the price of iron on the quantity in the market, he discovered that it was in fact positive -- that the more iron there was, the higher the price was. This led him to claim that he had "discovered a new type of demand curve with a positive slope".

His problem was one we studied extensively in our theory classes, but never discussed in the context of a concrete example, called the endogeneity of the regressors. Forget the fancy name -- the idea is simple. Think of the price as the intersection of a downward-sloping demand curve and an upward-sloping supply curve. If the system has no noise, you will only ever observe a single quantity and price -- the equilibrium. However, we can imagine that we observe variation because shocks move the demand or supply curve around. Now, suppose the demand stays fixed, and all the shocks move the supply curve. Then you will observe a series of points that describe the demand curve, and a regression will measure its (negative) slope. However, if the shocks only move the demand curve, you observe the supply curve's (positive) slope. In real life, you need to argue which of these cases -- or something in between -- your data describes if you want to do a simple regression of price on quantity.

(This is called endogeneity of the regressors because the error terms might affect the regressors -- in this case, your unobserved shocks may shift the quantity in a systematic way as well as the price by shifting the demand as well as the supply curve.)

One interesting thing is how this error can now be neatly and impeccably described in a page and a half of very clear statistical theory, but at the time it took thirteen years (until Working's 1927 paper) to discover the mistake. At the time, the idea of using statistics to measure and test economic theory was relatively young, but now our understanding has been refined to the point that an error made by one of the great founders of the field can be expressed in terms that an undergraduate could immediately grasp. It kind of makes you wonder what sort of boneheaded things we're all doing now that will seem trivial ninety years from now.

The Best Clubbing Experience of My Life.

noreply@blogger.com (thegio) — Fri, 12 Oct 2007 19:49:00 +0000

...was at 7pm this afternoon. Generally I hate clubs. To explain this, I often tell people that I don't like to dance. But that's not true -- I just don't like tepid bouncing in the dark to music I don't like.

But today I went mobile clubbing. These are flash mob events that are organized every few months. For mobile clubbing, you show up at a pre-specified public place with your personal music player. At the appointed time, someone somewhere gives a signal, and everyone puts on their headphones and starts dancing to their own music.

Today, it was in the Tate Modern museum's Turbine Room. There were hundreds of people -- little kids, young people, business (wo)men, elderly couples. Everyone was dancing to their own music, and many people were dancing really crazy. I know I was dancing really crazy. It looked very funny when you took your headphones off, because everyone was dancing wildly to no audible music. As far as I could see, nobody was drinking or acting drunk, unless you counted our hyperactive dance moves. Normal museum goes were startled, and I'm sure some joined in. It was, of course, impossible to tell who came specifically for the flash mob and who just joined in when they saw it going on.

We danced for an hour and a half and then left (Lori's ipod ran out of batteries). It was dwindling, but still going strong. I was covered in sweat, and LCD Soundsystem, Of Montreal, and Lucky Soul were echoing in my ears. This was everything dancing in public should be!