Forecasting and Tomorrow’s Jobs Report | Jared Bernstein | On the Economy

Jared Bernstein è un economista del lavoro statunitense ed è attualmente tra i consulenti di Barack Obama. Potete trovare qui una sua biografia un po’ meno sintetica.

Jared Bernstein

Ha un blog – On the Economy: Facts, Thoughts and Commentary by Jared Bernstein – da cui ho tratto l’articolo che segue (pubblicato il 5 luglio 2012, alla vigilia della pubblicazione dei dati di giugno sull’occupazione negli USA, e ripreso da

Forecasting and Tomorrow’s Jobs Report

I had a chat with a friend the other day – a prominent academic economist whose name I won’t disclose so he doesn’t get shunned in the faculty room – wherein we bemoaned the state of a) micro-theory (predicts implausible elasticities that never show up in the data; marginal product theory – a core premise – looking ever more suspect*) and b) macro-theory (a terrible muddle these days, as Paul K stresses).

But we agreed that econometrics still rules. Sure, there are those who practice eCONomeTRICKS, but “we regard them with scorn” (extra points for those who can source that quote without Google—even more points for those who can identify why it fits in an econometrics post).

I used to have decent econometrics – statistical analysis of economic data – chops, especially for a former musician/social worker, but alas, no more. I can still reliably run reduced form regressions and the Kalman Filter using the structural (or “state-space”) model I associate with Andrew Harvey (see previous link). But I simply haven’t kept up with the cutting edge stuff, though luckily, I know folks who have.

All of which brings me to the fool’s errand of forecasting employment growth for tomorrow’s jobs report. The consensus is for about 100K. I run a couple of models. At this point in the month, I run a regression of the log changes in payrolls on the lagged quarterly payroll growth, the monthly average of 4-wk UI claims, and the ADP (again, all in log changes) and forecast one month ahead (using the actual UI and ADP data for June).

I also try to tease out the longer term trend using the Kalman filter on the NSA data – this is a very good way to get at the underlying recent trend, which right now is running at around 90K, which is actually close to what I get with the standard time series regression noted above. So that’s about what I expect tomorrow, though given the confidence interval of 100K around these data along with the monthly revisions, the firm birth/death modeling – well, I don’t know anyone who has a great track record on this one.

However, that’s less a critique of econometrics than a warning about realistic expectations when forecasting high-frequency data.

Jared Bernstein

Poche mie considerazioni:

  1. Beati i cittadini di paesi (che non sono moltissimi, temo) in cui un consulente del governo può permettersi di avere un blog e di dire liberamente la sua “senza filtro” e senza doversi nascondere dietro un nom de plume.
  2. l’asterisco nel primo capoverso rinvia a questa gustosa nota:
    The great Joe Stiglitz gave a talk recently at the LSE on his new book on inequality (I also interviewed Joe the other day).  Anyway, a bit into the interview, he tells the LSE students, and I’m paraphrasing, “You know, that marginal product theory you’re learning around wage setting—it’s not true…you still have to learn it, but it doesn’t really work.”
  3. Paul K è chiaramente Paul Krugman.
  4. eCONomeTRICKS è un gioco di parole.
  5. Ho dovuto usare Google, ma “we regard them with scorn” è un verso della canzone The Folk Song Army di Tom Lehrer.
  6. Per la cronaca, il dato pubblicato oggi è + 80.000.

10 trucchi per sembrare meno stupidi

Puoi anche essere un genio, ma il rischio di fare la figura del cretino è in agguato.

Calvin Sun sul blog 10 Things ha pubblicato oggi (6 luglio 2012) un decalogo per evitare figuracce:

  1. Conoscere a menadito la propria materia
  2. Anticipare il proprio interlocutore
  3. Non fingere di sapere la risposta
  4. Giocare l’ignoranza a proprio vantaggio
  5. Mettere in luce quanto si sa e si è fatto
  6. Fare domande di conferma (e incorporarvi i dubbi)
  7. Includere le ipotesi e i limiti di applicabilità nella risposta
  8. Ricordarsi che «certamente» può ritorcersi contro di voi
  9. Dire piuttosto «Sarei sorpreso se…» (la risposta del Capitano Renault)
  10. Avere a portata di mano dati e citazioni rilevanti.
La cena dei cretini

10 things you can do to keep from looking stupid | TechRepublic

1: Know the material

As obvious as it sounds, nothing does more to prevent problems than knowing the subject you are discussing. The more you know, and the more insight you can provide based on your own experience, the less likelihood that you will misspeak or state an incorrect position. Even more important, knowing the material will give you confidence, and that confidence will show in the tone of your voice and in your body language. Do you know how to reduce the chances of being burned while working with a particular product? Don’t be afraid to share that knowledge.

2: Think three steps ahead of the other person(s)

This point relates to the first one. Not only must you know what you’re talking about, you also must anticipate the most likely questions you will get and prepare answers. In other words, you must do more than simply repeat information. You must be able to analyze it and show how it relates to the objectives and concerns of your listeners. If you are talking about a software implementation, what are the most likely areas where a problem will occur? What combination of hardware and software will be the most difficult to troubleshoot? If you have these answers, your listeners will appreciate your information more.

3: Don’t fake an answer

No matter how much you prepare, you might get a question for which you don’t know the answer. In such a case, resist the urge to guess. You might be right, but the chances are greater that you will be wrong, and an initial wrong answer followed by a correction will be worse than stating that you do not know the answer. Of course, if the question involves a complicated situation, people will be more understanding of your inability to answer than if you lack an answer to a basic question.

At the same time, try to answer what you can. If the question involves the interaction of multiple software products, for example, answer what you can about the individual products, then simply state that explaining the way they interact would take additional analysis.

4: Put a positive spin on lack of knowledge

Even though you might not know the answer, try to avoid saying so. Instead, try the old standard “That’s a good question.” Then explain the issues involved. If the answer will vary depending on different sets of circumstances or system configurations, you could talk about one specific circumstance or configuration and explain that one in detail. Then caution your listener that the results might be different in other circumstances.

5: Mention what steps you already took

Let’s say that you are a level one help desk analyst and you are escalating an issue to level two or beyond. When discussing the issue with the next analyst, make it clear what initial troubleshooting steps you already took and that they failed to work. If you don’t, that level two person might think that you neglected those steps and will think that you are incompetent. Better to be in front of the situation and explain what you already did than to have to react to the other analyst’s questions.

6: Incorporate alternatives when you ask a confirming question

If you are unclear about something you heard, incorporate into your question the possible alternatives. The person who is explaining might not be aware of those other alternatives and mistakenly believe your question is stupid.

For instance, suppose someone is explaining that a supplier is based in Arlington, and that person is aware of only the Arlington in Virginia. If you were to ask, “Do you mean Arlington, VA?” that person, and possibly others, might consider it a stupid question. If you instead ask, “Do you mean Arlington, Virginia, Texas, or Massachusetts?” you subtly make it clear that your question is not stupid at all. In the same way, rather than asking, “Do we need PowerPoint to run the presentation?” consider instead “Do we really need PowerPoint or just the viewer?”

7: Be clear in your answer about assumptions and limitations

Any answer you give will depend on specific facts and circumstances. Therefore, be clear about them, because in other cases the answer might be different.

For example, let’s say that you are vendor management person for your IT organization, and an issue has arisen with a vendor. Suppose someone in the organization asks you about the timeframe your company has in which to sue a vendor, and you know the answer. In giving it, you probably would want to qualify your answer to say, “In state X, the time limit to sue is y years, but in other states it might be different.”

8: Remember that “definitely [not]” can come back to haunt you

As soon as you say something “definitely” will or won’t happen, events will prove you wrong. As a result, you will end up with the proverbial egg on your face. A better alternative to “definitely will happen” is a response such as, “It might not happen, but the chances of that are really small.” An alternative to “definitely won’t happen,” might be, “It’s possible but extremely unlikely.”

9: Consider the Captain Renault “would be shocked” response

In the immortal movie Casablanca, Captain Renault declared that he was, “shocked, shocked I tell you” to find that gambling was occurring at Rick’s Café. You can use this dialog yourself to avoid looking foolish.

While the previous answers of “possible but unlikely” are better than the “definitely” or “definitely not,” they still carry an element of uncertainty. For that reason, my own preference is to answer so that the answer does have certainty. However, the certainty is not about the result, but about my reaction if the result is different. It also lets people know that you’re already aware that you might get egg on your face, so if you’re wrong, you don’t look quite as foolish.

So, for example, in response to the question “Does this Microsoft product have security issues?” I might answer, “If it doesn’t, I would be shocked.” If I am positive that a project will be late, I might say, “If this project comes in on time, I will be shocked.”

10:  Have data and citations in writing

If you are using data to support your points, have that data with you in writing or least have a citation to it. That way, you are not seen as making up numbers. Furthermore, people who disagree with you also have to disagree with data that came from someone other than you. Having the data and the citations gives you added credibility.

Nora Ephron e Lonesome George | The Economist

Ho già commemorato sia, su questo blog,  Nora Ephron, sia Lonesome George (su facebook).

Ora The Economist dedica loro un necrologio congiunto, a partire da un pretesto molto delicato. Vale la pena di leggerlo. Come le Vite parallele di Plutarco.

Nora Ephron

Nora Ephron and Lonesome George: When Nora met George | The Economist

THE first story Nora Ephron had to write for the New York Post – the one that made the guys on the city desk fall around laughing, got her the job, and launched her on a career of witty, wise writing on surviving modern life – was about a pair of hooded seals at the Coney Island aquarium. They were not only not mating, as they were supposed to, but also refusing to have anything to do with each other.

Lonesome George could relate to that. Though he was probably the last surviving example of the giant Galápagos tortoise Chelonoidis nigra abingdonii, he too refused to perform. Scientists, tourists, journalists, conservationists and the government of Ecuador all waited for two decades for him to mate successfully or, indeed, get it on at all. He wasn’t playing. In 1993 two females of a slightly different subspecies were put into his corral. He ignored them. When at last he decided to do his duty, in 2008 and occasionally later, the eggs failed to hatch. Clearly, he was a slow burner. Possibly he was gay. He refused to be turned on even when a female Swiss zoology graduate, smeared with tortoise hormones, gave him manual stimulation for four months.

At which stage Ms Ephron might have asked, couldn’t he at least have faked it? Women did that all the time. The most famous scene of her highly successful screenwriting career – which included “Sleepless in Seattle”, “You’ve Got Mail” and “Julie and Julia” – was the one in “When Harry Met Sally” (1989) in which Sally faked an orgasm over lunch at a deli on New York’s Lower East Side. After she had reared, moaned, gasped and shouted “Yes!” for what seemed like five minutes, the elderly lady at a nearby table told the waiter: “I’ll have what she’s having.”

George couldn’t. But then he didn’t have the advantage of living, like Ms Ephron, in Manhattan, where something was almost bound to happen to you if you simply stepped outside, and if it didn’t happen you could pick up the phone and order it. He was living in a volcanic field on Santa Cruz island 500 miles off the coast of Ecuador, where you couldn’t find a decent bagel if you tried: in fact, a place a lot like Washington.

Besides, he was no looker. Ms Ephron, though striking and svelte all her life, worried in the niggling way of women that her breasts were too small, her neck too crêpy (“I Feel Bad About My Neck” was the title of one book), her skin dry and her purse just wrong. George, whose neck was three feet of scrag and whose skin would have made several dozen purses, all thick, dry leather, didn’t care two hoots. His one concession to fashion was a shell in taupe. Ms Ephron preferred black; but taupe, especially on a couch, didn’t show the dirt.

Couches arose because they were part of the extra-domestic arrangements of her second husband, Carl Bernstein of the Washington Post. (Hence her Washington exile.) He left her when she was pregnant with their second child, falling for a woman whose neck was approximately as long as George’s and whose feet were splayed. Ms Ephron triumphantly got over it by turning the saga into a bestseller, “Heartburn”, in 1983, and then into a screenplay for the film, in which she was played by Meryl Streep. (“If your husband is cheating on you with a carhop, get Meryl Streep to play you. You’ll feel much better.”)

Her problem, she admitted, was falling in the first place for a priapic Jewish prince (Homo princeps judaicus incapax) – the sort who asked, “Where’s the butter?” when he meant, “Get me the butter.” She would hardly have done much better with George, whose neglected and expendable paramours were called No. 106 and No. 107, and whose rear-end shoves of his female companions were not, as scientists first thought, a playful invitation to sex but an order to quit the mud wallow, now, and leave it to him. The wallow was where El Solitario Jorge spent much of his time, smiling inanely and with the odd bit of food trailing from his mouth. All he needed was a copy of Architectural Digest to look like someone she knew.

To get her own back at Mr Bernstein, in a way befitting her love of food and her urge to cook, Ms Ephron in her book threw a key lime pie at him. She also provided the recipe. An erring tortoise would need sterner treatment. For tortoise soup, take one reptile, eviscerate, remove skin and all fat, remove shell. Simmer with lots of vegetables for two hours. Serves eight. This soup is illegal outside South Carolina.

And that was precisely how George had attained his solitary state: fishermen and sailors had eaten his tribe, goats had competed for the sparse vegetation, pigs had devoured the eggs, until he alone remained on Pinta Island, where a Hungarian looking for snails had discovered him in 1971. On his new island, he wasn’t truly alone: what with 180,000 tourists a year, 20,000 other tortoises, a team of scientists watching his every move, and journalists fighting on the bridge beside his pen. He was just the rarest creature in the world.

Equally, Ms Ephron – though she became the model of the defiant single woman – was not alone for long. She surrounded herself with blaring, cornucopian Manhattan, plunged into her work and then, in 1987, discovered the “secret to life” by marrying an Italian. But she was rare because, being very hurt, she got over it by making the world laugh at and with her.