November 30, 2007

Books to Read While the Algae Grow in Your Fur, November 2007

Richard A. Berk, Regression Analysis: A Constructive Critique
Presumes at least minimal familiarity with the math and practicalities of doing a multiple linear regression. The point of this book is to explain very clearly what regression can and cannot do, and especially to drive home the meaning and force of all the assumptions which are required to make the machinery of statistical inference for linear regression work. These assumptions are, or ought to be treated as, scientific hypotheses, which need to be not just taken for granted, or even listed mechanically and then ignored, but supported. (Many of them are hard to even assert with a straight face about important kinds of real data, never mind back up.) As he rightly says, the requirements for using regression for causal inference are even stronger, and the common practice of ignoring these issues, or hoping that they'll go away if you just use instrumental variables, has nothing to recommend it. (However, the discussion of Judea Pearl's work on causal inference in section 10.5 seems to me to be somewhat superficial, and even to misunderstand Pearl's book more than a little.) Of course, as descriptive summary of a data set, regression has much to recommend it; but not necessarily more than newer methods of data-mining, which he considers briefly in the conclusion.
I won't say this belongs on every practitioner's shelf, because it's not the kind of book you'll come back to again and again. I will say that anyone who is becoming a practitioner of regression needs to learn these lessons thoroughly, for their own sake and for the sake of anyone who might have to rely on their findings.
Errata: p. 66, the statistic oscillates between being S and being W, and the function Q is nowhere defined; p. 129, the first two terms in the numerator of eq. 8.9 should be grouped together in a parenthesis.
Terry Pratchett, Making Money
One of the great humanist writers of our time takes as his text Emerson's dictum that "Money, which represents the prose of life, and which is hardly spoken of in parlors without an apology, is, in its effects and laws, as beautiful as roses." (This line does not of course appear in the book.) Along with golems, eleven-dimensional cabinets of curiosities, hydraulic computers constructed by mad economists, etc., etc.
Christian Gouriéroux and Alain Monfort, Simulation-Based Econometric Methods
A book on a very important advance in connecting complicated simulation models to empirical data, relevant far more broadly than just in econometrics. The key trick, which I think is simply a stroke of genius, is the authors' idea of "indirect inference", which goes as follows. Find some "auxiliary" model for the kind of data you've got --- it doesn't have to be at all a good model, just one whose parameters are easy to estimate. Now fit the auxiliary to simulations from different settings of your real model, and to your actual data. Go with the value of the primary parameter whose auxiliary model looks most like the data's. In other words, use the model to predict what you get when you fit the auxiliary, instead of directly predicting what the data look like. This sounds bass-ackwards but it really works, provided the auxiliary model is sensitive enough to tell apart different parameter settings in the primary model. Unfortunately, they wrote this for hardened addicts of theoretical statistics and econometrics only.
Full review: By Indirection Find Direction Out.
Robert A. Dahl, Preface to Economic Democracy
An extended argument for introducing democratic decision-making inside economic enterprises, i.e., some form of democratic worker self-management and self-ownership.
The argument for economic democracy is presented as resting on a prior argument to a right to democratic self-government, that is, that one has a right to participate in deciding how one will be ruled, and in holding those with power of you accountable. This is a very basic form of liberty, which entails other liberties and rights if it to be effective. This given, Dahl then argues that there are enough power relations in a modern enterprise, even with the nominal "exit" option of quitting one job in the hope of finding another, that workplaces, too, should be governed democratically. I should add that he envisages this as being representative democracy, at least in enterprises of any size, and probably with professional managers hired by the workers and responsible to them. (Here, as in the larger polity, one advantage of representative over participatory democracy is that one can enjoy what Arendt somewhere calls "freedom from politics", or Constant's "liberty of the moderns", if one so chooses; I don't think Dahl mentions this point.)
After the consideration of rights, Dahl turns to the possible secondary costs and benefits: whether economic democracy would make workers better citizens, whether it would excessively reduce investment and innovation, etc.. He concludes, plausibly but of course by no means conclusively, that there is little reason to think it would have much immediate negative or positive effect on any of these matters. (As he points out, the circumstances under which it is now most common for a firm to become self-governing are when it's otherwise going to go under, and it's hardly surprising that worker ownership isn't a panacea for economic distress.) He does go so far as to express the hope, however, that the long-run effects of wide-spread economic democracy would be beneficial.
The issue which sticks in my mind about this is that of determining the limits of the self-governing unit. The natural guide here would seem to be power relations rather than physical contiguity --- the firm rather than the plant. But even so there could be all kinds of oddities. If, say, the senior management unit of a large firm got itself legally spun off as a separate corporation, which just so happened to have locked in long-term contracts to run all the other divisions, well, it would somewhat defeat the purpose to have those divisions and the management unit all be internally self-governing. To some extent Dahl's vision may have been superseded by events, such as the changes in corporate form and the rise of corporate networks which concentrate control even as they disperse ownership. This creates a problem if one finds his arguments on behalf of economic democracy appealing...
Nonetheless, I strongly recommend reading this: it is well-written, well-thought-out, well-argued, and short.
Alien
This seems like a much more connected story now than when I was ten or so.
Pluses: it's actually really rather creepy and atmospheric. (I particularly like the wordless, almost silent opening sequence.) It's also good that we don't spend too much time actually looking at the alien --- which doesn't seem cheesy yet, despite the advances of special effects and its innumerable cinematic progeny (raptors, anyone?). The crew's concern for the ship's cat is, also, a nice touch.
Minuses (including potential SPOILERS): The shower scene was cheesy. Ditto the alien slowly extending a tentacle up the inside of the legs of the other female character. The crew's general disregard for what seem like obvious safety protocols. (E.g., if an alien thing is sitting on your crew-mate's face and looks like it's going to be there for a while, train a camera on it. E.g., if said alien thing has just disappeared, and you go in to look for it, close the door behind you. E.g., do not wander off alone in search of the ship's cat.) Of course, the computers were much too implausibly primitive. Finally, it's hard for me to work out a scenario where "the Company" (1) knows about the alien and what kind of creature it is, and (2) knows where to find one but (3) needs to divert another mission to pick up a specimen in this way.
Still, like I said: creepy.

Books to Read While the Algae Grow in Your Fur; The Dismal Science; Scientifiction and Fantastica; Enigmas of Chance; The Progressive Forces

Posted at November 30, 2007 23:59 | permanent link

Three-Toed Sloth