At Play in the Fields of Data

I continue reading / skimming my way through Gelman and Hill's Data Analysis Using Regression and Multilevel/Hierarchical Models. I very much like the tone of the book. It is practical... not doctrinaire. Plenty of examples. R code where you need it. Confidence intervals are plus or minus 2 standard errors, not 1.96 or whatever Student's t requires. Think about scales and units and log transformations... Don't just think about them: Try them out! Mess around with the data. Make some plots comparing your confidence intervals. Questions of causation are yet to come, but I anticipate that Gelman and Hill are not structural purists, nor identification cops. They want you to think about your problem, know your data, and especially be aware of other related results.

Everything has been comfortably familiar until the chapter on simulation of probability models. Toto, we're not in Kansas anymore... welcome to the land of Bayes.

Mountaintop madness

Of all the creatively awful ways humans have figured out for messing up the natural environment, mountaintop removal coal mining may be the most appalling and senseless. Of course, we have managed to mess things up on a massive scale many times before: deforestation of large portions of whole continents is a good example. But amazingly, trees and forests grow back: witness the northeastern United States. Mountaintops do not grow back, except possibly on a geologic time scale. And adding insult to violent injury, the mining spoils kill the streams, and the coal itself has a second and even a third round opportunity to kill the planet: globally through CO2 emissions, and locally through coal ash disposal. I guess there are people who look at pictures like these and can only see dollar signs; others turn away and see votes. Now and then somebody stands up and does the right thing...

Winter Olympics Update

Viktor Ahn: Athlete of the games.

Winter Olympics Update

Women's figure skating: Very fine. I thought Kim Yu-na was better, but then, I can't tell a Lutz from a Salchow...
Women's half-pipe skiing: Less than half-worth watching, like any half-pipe event.

How big is it?

A question that comes up (or that should come up!) in empirical research is: How big is the effect? How important? How does this effect compare with that one? Deirdre McCloskey calls it the question of "oomph." For example, in explaining variation in earnings across individuals, which has more oomph: differences in gender, in education, or in work experience?

To answer, we need estimates of the partial effects, but also a way to scale the units to make comparisons between apples and oranges. One conventional way to do this is to standardize regression coefficients by calculating the effect of a one-standard-deviation change in each variable. But as I realized while teaching this to my econometrics students this week, the comparison is tricky when some of your explanatory variables are qualitative (0-1), such as gender. What does it mean to predict the effect of a standard deviation of female-ness? (Yeah, OK, maybe something, but still...)

During my midterm today I started reading Gelman and Hill's Data Analysis Using Regression and Multilevel/Hierarchical Models (they really could have used a catchier title!). It's interesting to read a first-rate non-economist statistician on the techniques we economists use routinely. I have already gained one tip that helps with the problem at hand- the problem of oomph. Whereas standardized coefficients usually look at the effect of a one standard deviation change in X, Gelman recommends scaling regression coefficients by two s.d. Why? This makes comparison with the 0-1 effects of dummy variables more reasonable. When the mean of a dummy variable is around 0.5 (e.g., female), then one s.d. of it is also 0.5, so a 0-1 switch is 2 standard deviations of the dummy. And it's pretty close even when p = 0.2: s.d. = 0.4. Voila! I can compare the size of the effect of gender on earnings with the effect of experience or education.

Nifty! And practical! And easy... And did I mention that all their examples are cleanly coded in R? That's especially handy for ECON 41/42 at Santa Clara University. It's a fat book, and it gets harder, but I'll keep reading.

Winter Olympics Update

Ice dancing: Gorgeous!
Interview of Bode Miller: Shameful
Aerials skiing: Yawn
Curling: Haven't seen it, haven't missed it
Caddy ELR commercial: The perfect 60-second commercial; narrative, pacing, acting, cinematography, music, and theme all come together as they rarely do in 2-hour movies. Did I mention that it is almost insufferably obnoxious? The ironic tone does not cancel out the self-satisfied jingoistic swagger of the thing. But hey, if it gets bros with dough to consider a plug-in hybrid, I'm all for it...

India's thrifty mission to Mars

Now if only we could convince them to send Sandra Bullock along...

The budget of India’s Mars mission, by contrast, was just three-quarters of the $100 million that Hollywood spent on last year’s space-based hit, “Gravity.”

Umiak kayak song redux

Three years ago I posted the lyrics I remembered from a grade school song in which one learns a few "eskimo words." I now realize that this post has garnered a number of comments over the intervening period. At the time, I promised to follow up with a post of the musical notes for the melody, which I then shamefully forgot to do. Anyway, here's a link to a midi based on what I recollect of the cheesy but catchy "Native American"-style melody, in the key of A minor. This only goes through the tricky "mukluk" verse... you can improvise the balance in the same style. Don't forget to include a dramatic ritardando and hushed pianissimo for the two lines just before the booming "Learn them if you can!" ending: a tempo, fortississimo!

Melody here.

Lyrics:

Umiak, kayak, mukluk, tupik,
Umiak, kayak, mukluk, tupik,
Umiak, kayak, Eskimo words,
Learn them if you can.

Umiak, a boat for many men,
Kayak, a boat for one man,
Umiak, kayak, Eskimo words,
Learn them if you can.

Mukluk, an Eskimo boot,
Tupik, an Eskimo tent.
If you heard an Eskimo say these words,
You'd know exactly what he meant.

Umiak, kayak, mukluk, tupik,
Umiak, kayak, mukluk, tupik,
Umiak, kayak,
Eskimo words,
Learn them if you can.

How close are we to full employment?

One of the most striking--and depressing--aspects of the great recession has been the collapse of the employment-population (E/P) ratio and its stubborn failure to recover. Unlike the conventional unemployment rate, which does not count discouraged workers as unemployed if they stop looking for work, the E/P accounts for discouraged workers and is thus often considered a superior indicator of aggregate labor-market conditions. By this metric, the recovery has a long way to go.

Or does it? E/P is experiencing an underlying downward trend as well as the cyclical effect of the recession. The trend change is due to falling labor-force participation, which in turn reflects changing demographics and to some extent changes in policies such as disability insurance. If the downward trend were strong enough, E/P might actually be a lot closer to "full employment" than it appears to be.

This post summarizes some recent work on the question, coming out of Federal Reserve banks. A key challenge is that the underlying "trend" E/P is not observed, but must be inferred; the assumptions that go into that inference matter. Furthermore, and perhaps less tractable, participation may be subject to hysteresis, meaning the trend may actually have been shifted by the depressed conditions of the recent period. If so, the past relationship between demographic characteristics and work behavior is not a reliable guide to the current relationship.

My read is that labor markets are not yet close to full employment, but the full-employment target is hard to locate with much precision. Giving the benefit of the doubt to the workers, at the very least monetary policy should remain loose until we see some serious evidence of wage inflation.

Winter Olympics Update

Men's figure skating: snoozefest, falls, bad outfits.
Bobsledding: Perennially boring.
Super G: Super.
Short-track speed skating: Best winter olympic sport ever.
Ice dancing short program: Ella and Louis... need I say more?
Charmin Zamboni bathroom break commercial: Outstanding.

Alambique Flat...

... is a lovely grove of second-growth redwoods in Wunderlich County Park. I have walked this park perhaps more than any other. You can take most of it in with a five-mile circuit. Alambique Flat is about as far as you can get from the trailhead, and on a weekday you can have this supremely peaceful place all to yourself. The grove backs up to a burbling creek. The opposite hillside appears to be a drooping mudslide of sodden earth, but in fact it is a solid piece of something like flowstone... might there be caverns below?

While investigating the limestone in the creekbed, you might come across this fallen log covered with a troop of tiny ghostly soldiers, fruiting bodies of a slime mold.

Nearby, "The Meadows." By now the (non-native) grasses should be lush. This year? Well, they're trying. Maybe these clouds will bring them something they can use.

The banality of NSA misbehavior

So it seems increasingly clear that a healthy percentage of the NSA's spying activities are not directed at protecting us from nasty terrorists, but amount to some form or other of industrial espionage, implicitly if not explicitly on behalf of U.S. corporations.

... disclosures in recent months from the documents leaked by Mr. Snowden show the agency routinely spies on trade negotiations, communications of economic officials in other countries and even foreign corporations. 

American intelligence officials do not deny that they collect economic information from overseas, but argue that they do not engage in industrial espionage by sharing that information with American businesses. China, for example, is often accused of stealing business secrets from Western corporations and passing them to Chinese corporations.

And why shouldn't I believe you?!

Workers reject UAW in TN VW plant vote

Some years ago a dude asked, Why is there no socialism in the United States? Not yet, but the Republicans are trying so hard to make it happen...

John Wright, 43, is a test driver at the plant and identifies as a right-leaning independent. He says he makes between $30,000 and $40,000 a year, and supports a wife and three young daughters. When Corker  -- who takes more money from the securities and investment industry than any other -- came back to Nashville to voice his opposition to the UAW, Wright was puzzled. 

"It made me really start thinking about my position, when it comes to political parties, because I can't for the life of me understand why the Republicans and big money are coming against us so bad. To me, they're attacking the average worker," Wright said, in the hours before the election results were announced. "To have politicians think that there's nothing more important than coming down and picking on the little guy because he wants a union, there's a national debt we've got to control, we have foreign policy things that we elect them to go up there to do, but you have to fly home for an emergency meeting because I want a union?"

Can Music Be Perfect? Vol. 39

Speaking of "You Stepped Out of a Dream," here's the best rendition ever. Sax and bass. Braxton plays it pretty straight, but with enough edge to remind you who's playing. Dave Holland offers an a cappella bass solo for the ages. I have this album in vinyl, and it is brilliant start to finish. Apparently still unavailable on CD, unless you want to shell out $300 for Mosaic's Braxton box set. But you should buy the vinyl anyway, if only for the extra cool album cover.

You Stepped Out of a Dream

The whole thing is excellent, but esp. around 16:00 when Dexter steps in. Every note he plays counts. Oscar is humming and sweating, enjoying himself.

Your recipe of the day

Faber's Okra Soup.
Put on to boil early a shin of beef; after it boils put in 4 quarts of okra cut up, one 3-pound can of tomatoes. Boil for hours.
- Carolina Rice Cook Book (1901)

Meeting at 7:30 a.m. tomorrow

Oof... if only I could be more like Captain Renault...

Dean Baker on drug patents

He is a bit of a broken record on this topic, but then again, it is important, and his latest post is particularly nicely done. Anyone care to offer a counterargument?

Winter Olympics Update

Moguls skiing: boring as shit.
Ice dancing: better than I remember.
Snowboarding slopestyle: unfuckingbelievable!
That is all for now.

Man kann nicht nur auf Punk leben

In honor of the Olympics, a couple of Russians playing music...

C'mon

Oh Come On

Against all odds, punk lives!

Still my favorite Philip Seymour Hoffman part

Back before he learned to overact, and back when PT Anderson was making really good movies.

Marmot Day

Our California version of the groundhog is the delightful yellow-bellied marmot. I hope they see their shadow tomorrow, because we sure as hell need more winter out here... as in precipitation.