Pictures.

I post a lot of pictures on Facebook, but it occurs to me that most people who don't actually live with me can't see them. So here are a handful of pictures of me in London.

Menomena

Could this be where the very nice band Menomena got its name?

What Is It I Study Again?

When introduced to new people, I am usually asked what I study. And even after eight months I still relish the awkward pause that inevitably follows my answer, "econometrics", as people struggle to think of something they can say about econometrics or at least remember what it is.

So what is it? I am not completely convinced I know the answer definitively myself. It is in a fuzzy area between economics and statistics. And I'm maybe not alone in note knowing exactly where the boundaries lie: I recently was pointed to an interesting discussion about the difference between econometrics and applied statistics.

Actually, reading this, I think my taste and interests are probably more in line with the statisticians than the econometricians.

I Inadvertently Give America a Bad Reputation on Global Warming.

One of my flatmates remarked that she believed that the unusually cool weather in London last week was because of global warming. I said that although I that might be true, we don't really understand how global warming will affect local weather conditions well enough to make assertions like that. She countered by claiming that the weather in London this winter was so abnormal as to constitute strong evidence for global warming. I said that I suspected it was well within the bounds of normal random weather behavior (and that's what prompted me to look at the data that led to my last post).

I learned today that when she told her roommate that I had said this, her roommate responded by saying, "that's just the response that you should expect from an American." It turns out that she interpreted and passed on my objection as saying that I didn't believe in global warming. And although I repeatedly told her that I believed it was a problem, and that there was good evidence pointing us towards the danger of global warming, unusual weather for a week -- or for a year, or for ten years -- in London did not constitute good evidence. I showed her my graph of historical temperatures where you can see clearly that although the weather has been unusual for the last ten years, it has been as "unusual" many times before, including many times before the industrial revolution when it was certainly not the effect of human carbon output.

It was odd how unable she seemed to decouple the two ideas. She kept challenging me with questions like, "don't you think that global warming would cause temperatures to rise?" "Don't you think we are changing the environment with pollutants?" And most significantly, "don't you think global warming will cause unpredictable weather behavior?" The answers are yes, yes, and yes, but that doesn't mean that everything that could be explained by global warming is evidence for it, especially in light of the last question -- the effects will be unpredictable, especially in something like weather, with a very high noise-to-signal ratio.

I'm not very sure I made her understand this even in the end. I'm afraid she's going to continue going around telling people about the American in her flat who doesn't believe in global warming.

By the way, although my test for a structural break in average temperature gave crazy results (suggesting that modeling weather as a normal distribution with a mean that changes reasonably smoothly is a flawed model), there is a statistically significant warming trend in the data that I have. It cannot be observed in only a handful of the data, but in a regression over all the recorded temperature for England, it appears that things have been getting gradually warmer in a way that is inconsistent with simple random chance. Again, this is not direct evidence for global warming per se, but it is more relevant than a single cool, rainy week in May.

Library Rant.

There is no study room in my dorm. I have a hard time studying in my room. In my kitchen, my flatmates come in and out. Sometimes I sit in the stairwell to study, but there's no desk there. I could walk forty-five minutes to LSE for a quiet place, but now that students are revising for exams, it's impossible to find a place in the library even there. There are a couple public libraries within twenty minute's walk, but they're noisy and have short opening hours. So when I found that King's College, a member of the University of London (the umbrella university that contains LSE), had a large, quiet library only ten minutes walk from my dorm, I was naturally excited.

When I went in, I asked how I could get access. They passed me from person to person, but finally a woman assured me that all I had to do was get my department to stamp some paperwork confirming that I was an LSE student, and I could get a permanent card. A couple days later I took the form to LSE, but when I read the description on the form more carefully, I found that it needed a stamp from a King's College department confirming that I was taking a course from them. Of course, I was not, so I checked their website. On it I found another form for access that did not require a department stamp. Wanting to be sure I had the right one before I left LSE that day, I called them.

The first woman I spoke with said that I actually needed a different form that required me to justify why I needed the library. I said that this wasn't like the other two forms, and asked that she check to make sure she was talking about the right one. She then passed me off to her supervisor, who asked me why I needed the library. I told him that I just wanted a quiet place to study. He said that wasn't good enough. "But King's College students can use our library," I objected. He denied this.

So I hung up and went across the street to our library and asked the reception. Sure enough, any King's student can come and use the library, without a department stamp or research justification, only because they are students at King's. So I called the first guy back and told him this. He said that I was mistaken. I said I was standing in front of the woman who decides who goes in and out, and she lets the in. He said that she was mistaken. I asked for his supervisor's name. He refused to tell me, and the conversation ended badly.

The woman at the LSE library admissions desk then gave me the number of her supervisor, who she said would be interested to hear about this. So I called them, and the woman I spoke to sounded very concerned. She said she would pass it on to her supervisor, and gave me his name. I waited for a few days and didn't hear anything. Then I sent him an email. Then I didn't hear anything for another couple days. Then I called again. The same woman I spoke to before said he would answer my email the next day. Two days later (today) I did hear from him, and he confirmed that I am not allowed to use the King's library except in the case of a research need, and even then, I am only allowed to look at or photocopy a book, not check it out.

This is infuriating. I am a member of the same university, and I can't even go into their library? At the same time, I can't find a space to study in my own library, and their students are granted unlimited access? And it took two weeks for the bureaucracy to even decide this conclusively? Is this what I pay thirty-two thousand dollars a year for?

I Shouldn't Be Surprised, But I Am a Little.

Basically, this year I learned statistics. And everything I learned was very abstract. In my classes, I was never once called upon to analyze an actual dataset. I did a little on my own as part of my effort to do all the homework for a simpler econometrics course, but it was all kind of canned and not very exciting.

But yesterday I got tired of studying book stuff, and wanted to do something a little more fun but still relevant, so I thought I would check for unit roots and structural breaks in weather data. (I just found out that you can get monthly temperature data for England going back to 1659! It's free at http://hadobs.metoffice.com/hadcet/data/download.html.) I wrote a program in gauss to do these things, and because I don't trust myself so much, I thought I would do some Monte Carlo trials to see if my statistics were being calculated correctly.

And the thing that shouldn't be surprising, but kind of is, is that these techniques really work! For example, I generated five hundred data according to this model:

y_{t+1}=2 + a*y_{t} + e

where e is a normally distributed random error term with mean zero and variance 0.25. I then set a=1 (that is, there is a unit root), and calculated the appropriate Dickey-Fuller test statistic. This rejects the presence of a unit root at 90% if the critical value is less than -2.57. In thirty trials, it came up with these statistics:

-0.43573590
-3.5956646 *
0.16992838
0.073596016
-2.6977168 *
1.8725950
-1.1551782
1.4518842
-0.72521093
0.88719054
0.19026298
0.74761789
-0.87596782
0.55484388
0.55116807
3.8321580
1.1601078
-2.1123924
-3.9787004 *
0.22736827
-1.3979254
2.0005073
-1.5957790
-2.6886918 *
-1.9549079
2.9131591
-0.76821185
0.76051801
-1.0562308
-0.27190441

Sure enough, it rejected incorrectly four times out of thirty, which is close enough to 10% for random data. Now, observe what happens if I change a to 0.999 -- that is, awfully close to one, but not quite a unit root. Here's what comes out:

-19.039612
-19.276780
-25.051717
-20.740252
-23.782902
-24.197974
-19.147238
-22.555892
-19.436182
-23.419553
-20.569042
-23.110684
-24.328111
-21.205045
-24.545090
-20.042392
-20.988465
-22.970617
-22.802407
-22.611619
-19.959404
-24.710798
-19.793600
-25.165184
-22.078757
-20.559708
-23.508961
-25.047222
-19.818685
-24.183677

100% rejection! This is a mighty powerful test. My structural break F-statistics are also impressive. It makes me have a little more faith that all this theory is really worth something after all.