Archive for September, 2010

What is a Kalman Filter?

September 28, 2010

Theoretically, the Kalman Filter is an estimator for what is called the linear-quadratic problem, which is the problem of estimating the instantaneous “state” […] of a linear system perturbed by white noise by using measurements linearly related to the state but corrupted by white noise. The resulting estimator is statistically optimal with respect to any quadratic function of estimation error.

Practically, it is certainly one of the greater discoveries in the history of statistical estimation theory and possibly the greatest discovery in the twentieth century. It has enabled humankind to do many things that could not have been done without it, and it has become as indispensable as silicon in the makeup of many electronic systems. Its most immediate applications have been for the control of complex dynamic systems such as continuous manufacturing processes, aircraft, ships, or spacecraft. To control a dynamic system, you must first know what it is doing. For these applications, it is not always possible or desirable to measure every variable that you want to control, and the Kalman filter provides a means for inferring the missing information from indirect (and noisy) measurements. The Kalman filter is also used for prediction the likely future courses of dynamic systems that people are not likely to control, such as the flow of rivers during flood, the trajectories of celestial bodies, or the prices of traded commodities.

Yay! The above is the opening paragraphs of Mohinder S. Grewal and Angus P. Andrews Kalman Filtering: Theory and Practice Using MATLAB, Second Edition (2001, John Wiley & Sons, Inc.), which currently is under my scrutiny.

Advertisements

Kuhn vs. Popper by Steve Fuller, Part 1

September 25, 2010

In Kuhn vs. Popper, Steve Fuller discusses and compares the two most important philosophers of science in the 20th century; Thomas Kuhn and Karl Popper. The book is not what I expected it to be (what I expected is not entirely clear, but that is not the point). The book as a broader view than I expected, for example, revisiting philosophers from Plato to contemporary and, to me, new names like Rorty, and shuffles over themes from theology to nazism. The broad view is indeed intentional and announced in the introduction:

This book is designed to recapture the full range of issues that separate these two self-styled defenders of science [Kuhn and Popper, of course].  Many of the issues plumb the depths of the Western psyche: What is the relationship beteween knowledge and power? Can science bring unity to knowledge? Can history bring meaning to life? At the same time, these issues are entangled in more secular concerns about economy and society, politics and war  most of which are still very much with us today [p. 3].*

While I cannot say I found every twist of the book equally interesting, it did help me understand the the ideas of Kuhn and Popper better than before and even correct things I got wrong. For one thing, I used to think of Popper as the normative and Kuhn as the descriptive; Fuller claims I was wrong:

It was not, as is often said, that Kuhn was more ‘descriptive’ and his rivals [the Popperians] more ‘prescriptive’ with respect to the history of science [p. 208].


Thomas Kuhn (1922 - 1996)

According to Fuller, the Popperians appeared ‘perversly contrarian’ to the public when they imposed a normative perspective on the history of science; to the public, science’s authority was self-evicent. Kuhn, on the other hand, ‘never articulated the norm’ he imposed when he selected and arranged the historical examples he used in his arguing (p. 209). In other words, one needs a rather good idea of the history of science in order to see and understand that Kuhn presents only the examples that fits with his theory; according to Fuller, hsitory is abound with examples less in accordance with Kuhn’s theory.

A thing I do not like about the book is that it is densely written, and in order to follow it, one need to at least be acquainted with the theories of Kuhn and Popper up front; Fuller only briefly summarizes the ideas. I find this surprising, as the cover is littered with acclaim from newspapers like the Economist and the Financial Times with the obvious purpose to attract the general reader. According to Fuller, this involved writing, presupposing knowledge of the issue at hand was a trademark of Popper, and, I’m afraid, many philosophers:

Popper and Adorno [another philosopher Popper locked horns with] shared the critic’s tendency [not entirely clear to me what ‘the critic’s tendency’ is supposed to point to here] to presuppose that the audience already knows the target of criticism in some detail, so that one’s own discourse becomes a series of reflections on the hidden opponent. This feature made it frustrating for listeners who sought  constructive advice on the conduct of social research [p. 154].

Frustrating indeed, do you listen, Fuller? (I guess not.)

Now, while flipping through the book, I realize I’ve underlined too many quotable passages and marked too many passages to address them all in one post. I will thus try to focus my opinion of this book in a later post; stay tuned.

* To be sure, page numbers refer to the Icon Books 2006 paperback edition.

Related post:

Picture of the Day: Rocket Science

September 22, 2010

Today’s picture is a tounge-in-cheek response to John Whitehead’s picture over at Env-Econ:

Behavioral Economics and the Environment

September 15, 2010

Gardner Brown and Daniel A. Hagen guestedited a recent special issue of Environmental & Resource Economics (Vol. 46, No. 2), and suprisingly wrote the first article themselves. They begin like this:

Many economists have embraced a paradigm characterized by perfect information, rational expectations and an otherwise benign environment in which perfect competition reigns, with very minor asides for imperfect competition. Rumblings of opposition have been growing louder. Nobel prizes are being awarded to scholars who have taken us out of this historic straight jacket. Lo and behold there can be asymmetric information, increasing returns to scale, cooperative behavior and agents who consistently fail to optimize.
Behavioral economics is another theme gathering strength, and it may be particularly germane to environmental and resource economics. Consumer theory, measurement of benefits, intergenerational discounting, mechanism design, and the role of fairness are all subjects of importance for environmental and resource economics and they are all areas in which behavioral economicsmay provide important insights.

A Dictionary of the Near Future

September 13, 2010

I am supposed to be working, but I then I an op-ed, by Doublas Coupland in The New York Times, caught my attention: A Dictionary of the Near Future:

The thing about the future is that it never feels the way we thought it would. New sensations require new terms; below are a few such terms to encapsulate our present moment.

I regcognize myself in a lot of the terms, Airport-Induced Identity Dysphoria, for example:

AIRPORT-INDUCED IDENTITY DYSPHORIA Describes the extent to which modern travel strips the traveler of just enough sense of identity so as to create a need to purchase stickers and gift knick-knacks that bolster their sense of slightly eroded personhood: flags of the world, family crests, school and university merchandise.

It goes deep, wonder what economists have to say about this:

CRYSTALLOGRAPHIC MONEY THEORY The hypothesis that money is a crystallization or condensation of time and free will, the two characteristics that separate humans from other species.

I am dimanchophobic every now and then, approximately once a week:

DIMANCHOPHOBIA Fear of Sundays, a condition that reflects fear of unstructured time. Also known as acalendrical anxiety. Not to be confused with didominicaphobia or kyriakephobia, fear of the Lord’s Day.

So not true:

INTRAVINCULAR FAMILIAL SILENCE We need to be around our families not because we have so many shared experiences to talk about, but because they know precisely which subjects to avoid.

Fair enough, true, but is it a real problem, or just the manifestation of deeper problems regarding attention spans or commitments, or both?

KARAOKEAL AMNESIA Most people don’t know the complete lyrics to almost any song, particularly the ones they hold most dear. (See also Lyrical Putty)

?:

PROCELERATION The acceleration of acceleration.

Mere word play:

PSEUDOALIENATION The inability of humans to create genuinely alienating situations. Anything made by humans is a de facto expression of humanity. Technology cannot be alienating because humans created it. Genuinely alien technologies can be created only by aliens. Technically, a situation one might describe as alienating is, in fact, “humanating.”

So that’s what standard deviation means, I hear it all the time:

STANDARD DEVIATION Feeling unique is no indication of uniqueness, and yet it is the feeling of uniqueness that convinces us we have souls.

Dense Philosophy of Science: A Necessity?

September 13, 2010

What we call the ‘modern’, and distinctly Western, sensibility emerged as people tried to organise the conduct of the sciences in light of second-order condsiderations of what might be common to all the sciences. The result was a Galilean zeal for spotting latent contradictions between bodies of knowledge, the pretext for eliminating the social, lingusitic and practical barriers to their proper integration into one system of thought. Popper promoted a version of this strategy in his attack on the ‘myth of the framework’, the Kuhnian idea that the presence of incommensuarable theories rendered any explicit normative comparison so difficult that one simply had to wait for history to take its course, as individuals come to adopt one or another theory for their own reasons. In contrast, Popper argued that if the incommensurable theories are truly scientific, they aspire to universality, which means that there will be cases that they have yet to explain or predict. These cases may then serve as relatively neutral ground for designing a crucial experiment to decide amonst the theories.*

I’m reading Steve Fuller’s truly interesting account of ‘The Struggle for the Soul of Science,’ the legendary debate between Kuhn and Popper. Although I enjoy the book, I cannot help but marvel at the dense style Fuller, and many in his field, pledge to. I wrote an essay once, on what I’ve late learned is known as the underdetermination hypothesis (data can be explained by any number of mutually incompatible theories, see, f.eks. pp. 61-62 in Kuhn vs. Popper). I wrote rather dense myself, I seem to remember (I’ll post an example if I can dig it up). I wonder if a complicated style is necessary to philosophy of science, if it is impossible to iron it out more, if it’s somehow inherent to the subject. Philosophy of science is, in a weird way, depending on itself for its own existence. (See; I came up with something rather dense just writing about how dense it is!)

* Steve Fuller (2003), Kuhn vs. Popper, pp. 66-67, Icon Books 2006 edition.

Superfreakonomics by Steven D. Levitt and Stephen J. Dubner

September 10, 2010

After having it on my shelf for quite a while, I finally sat down and read Superfreakonomics; Levitt and Dubner’s follow-up to their bestselling book Freakonomics. Superfreakonomics is laid out much the same way as Freakonomics was, although less time is spent on declearing Levitt to be a genius.  However, with chapter titles as How is a street prostitute like a department-store santa?, Why should suicide bombers buy life insurance?, and What do Al Gore and Mount Pinatubo have in common?, the similarity to Freakonomics is unmistakeable. The similarity also makes Superfreakonomics feel like an act of duty more than a work of inspiration.

Just like John Whitehead, I enjoyed Freakonomics more than I did Superfreakonomics. I also agree with Whitehead that the highlight is the epilog on monkeys learning to use money. Levitt and Dubner do a great job, however, coming up with surprising conclusions:

This is a strange twist. Many of the best and brightest womenin the United States get an MBA so they can earn high wages, but they end up marrying the best and brightest men, who also earn high wages which affords these women the luxury of not having to work so much (p. 46).*

So, perhaps there’s more to getting an MBA than high wages? Bright men, for example. Next, do what you want to do:

Deliberate practice has three key components: setting specific goals; obtaining immediate feedback; and concentrating as much on technique as on outcome. The people who become excellent at a given thing aren’t necessarily the same ones who seemed to be “gifted” at a young age. This suggest that when it comes to choosing a life path, people should do what they love […] because if you don’t love what you’re doing, you are unlikely to work hard enough to get very good at it (p. 61).

Partly beg to differ. I think many of those really good at something (like, world-class-good), at least has to begin practicing at an early age.

Other amusing and at times unsettling conclusions are that death rates in Los Angeles drop when doctors go on strike (p. 81), in Singapore they have the the Manitenance of Parents Act (p. 106), economists believe more in theory than in the real world (“Sure, it works in practice, but does it work in theory?”, p. 115), the Endangered Species Act endanger rather than protect species (p. 139), buying locally produced food increases greenhouse-gas emissions (p. 167), and the movement to stop global warming has taken on the feel of a religion (p. 169).

Perhaps the most controversial part of the book is the chapter on global warming, where Levitt and Dubner embrace geoengineering as the short-term solution. The noise around the chapter was seemingly so annoying to someone that critical posts on the Freakonomics blog were removed (see here, for example). Among the disturbing claims Levitt and Dubner provide is that climate scientists ‘turn their knobs’ such that their model do not provide outlier estimates, because an outlier model is hard to get funded. The economic reality of research funding generate a scientific consensus, rather than independent research (p. 182). The claim ressonates with a seminar I recently attended. The seminar was given by a Danish climate scientist who were concerned that the famous hockey stick graph, hailed as the undisputable proof of man-made climate change, resulted from lack of data, inappropriate methods, and an assumed stable temperature prior to the industiral revolution. The discussion in Superfreakonomics do, however, seem fairly balanced in places, see for example the discussion on page 199.

I would recommend Superfreakonomics to anyone unfamiliar with Freakonomics, but, honestly, it’s a Freakonomics 2, and not any more super than it’s predecessor, which is, notwithstanding, quite superb. Superfreakonomics is perhaps an easier read (or I’ve become a better and quicker reader), but lacks a character like Sudhir Venkatesh.

* Page numbers refer to the Allen Lane UK edition.

Dancing with Professors: An Essay on Academic Prose & Rhetoric by Patricia Nelson Limerick

September 7, 2010

Robert Kozinetz gave a really interesting talk in Bodø last week; he talked about how to develop ideas and how to make them matter. During the talk, he brought up Dancing with Professors; an article from the New York Times (I think it was) discussing troublesome academic prose. The article is written by Patricia Nelson Limerick and is from way back (2001, perhaps), but is still relevant, of course (old habits die hard, I guess).

While we waste our time fighting over ideological conformity in the scholarly world, horrible writing remains a far more important problem. For all their differences, most right-wing scholars and most left-wing scholars share a common allegiance to a cult of obscurity. Left, right and center all hide behind the idea that unintelligible prose indicates a sophisticated mind. The politically correct and the politically incorrect come together in the violence they commit against the English language.

The dancing comes in when Limerick claims (perhaps rightfully so) that those who become professors are those nobody wanted to dance with in high school; you know, the shy, fearful, and lonely guy in the corner:

Professors are often shy, timid and fearful people, and under those circumstances, dull, difficult prose can function as a kind of protective camouflage. When you write typical academic prose, it is nearly impossible to make a strong, clear statement. The benefit here is that no one can attack your position, say you are wrong or even raise questions about the accuracy of what you have said, if they cannot tell what you have said. In those terms, awful, indecipherable prose is its own form of armor, protecting the fragile, sensitive thoughts of timid souls.

After a couple of (dreary) parables and sidetracks, Limerick returns to how academic prose is hindered from improvement: Professors think they are supposed to teach bad writing in grad school:

This is a very well-established pattern, and it is the ruination of scholarly activity in the modern world. Many professors who teach graduate students think that one of their principal duties is to train students in the conventions of academic writing.I do not believe that professors enforce a standard of dull writing on graduate students in order to be cruel. They demand dreariness because they think that dreariness is in the students’ best interests. Professors believe that a dull writing style is an academic survival skill because they think that is what editors want, both editors of academic journals and editors of university presses. What we have here is a chain of misinformation and misunderstanding, where everyone thinks that the other guy is the one who demands, dull, impersonal prose.

The lesson? Think more like a carpenter than, say, an artist:

Ego is, of course, the key obstacle here. As badly as most of them write, professors are nonetheless proud and sensitive writers, resistant in criticism. But even the most desperate cases can be redeemed and persuaded to think of writing as a challenging craft, not as existential trauma. A few years ago, I began to look at carpenters and other artisans as the emotional model for writers. A carpenter, let us say, makes a door for a cabinet. If the door does not hang straight, the carpenter does not say, “I will not change that door; it is an expression of my individuality; who cares if it will not close?” Instead, the carpenter removes the door and works on it until it fits. That attitude, applied to writing, could be our salvation. If we thought more like carpenters, academic writers could find a route out of the trap of ego and vanity. Escaped from that trap, we could simply work on successive drafts until what we have to say is clear.

The Turbulent Side of Biodiversity

September 5, 2010

It’s been a while since I visited the Freakonomics blog. A while back, however, I started reading last year’s Superfreakonomics, and paid the blog a visit. My attention was attracted to a post on biodiversity, perhaps since I’ve lately taken an interest in the economics of biodiversity. The post is written by James McWilliams and asks:

[…] But it’s worth asking: what are we really talking about when we talk about biodiversity? On the surface, the word signifies the entirety of biological life. […] But not unlike the terms “environmentalism” and “sustainability,” biodiversity has a turbulent side, one with hidden implications that complicate its value as a precise gauge for land conservation.

Entirety of biological life? Hm, not sure if I agree.

The heroic efforts of ecologists notwithstanding—biodiversity remains an impossible concept to quantify in absolute terms. […] But critical questions remain: Is this erosion anything new in absolute terms? Is the decline in diversity that we’ve diligently documented and rightfully scorned reflective of genetic erosion as a whole? From the perspective of global biodiversity, does [for example] a salamander really matter?

A wise man once pointed out to me that much of human development and economic growth has relied on replacing ecosystems with monocultures (think clearing a forest to sow grains); a development impossible without loss of biodiverstiy. Notwithstanding:

One school of ecological thought rests on the premise that “biodiversity often peaks” in ecosystems that have been moderately disturbed by human development.

A final concern:

A final concern deals with the fact that, as we expand the built environment, some species will suffer the consequences while others will thrive, or at least suffer less. All of which raises a thorny philosophical question: who are we to decide which species deserve to flourish or suffer more than other species? Given that any sort of development, however aggressive, has the potential to influence an innumerable range of species in innumerable ways, we’re stuck with the task of somehow assigning comparative worth to plants and animals that have far outdated our own existence on the planet […]

Preserving and fostering biodiversity is a profoundly important environmental challenge, one that will only intensify throughout the century. But because the concept is so difficult to pin down and quantify, preserving it may require doing so through less expansive standards. More general, and policy-applicable, standards such as density of production, extent of open space, public health concerns, and the integration of built and natural environment might serve environmental concerns more efficiently than a concept that, theoretically speaking, has as much sympathy for a landfill as it does a rain forest.

National Research School 2010

September 1, 2010

The rest of the week, I’ll attend (and serve as discussant at) the so-called NFB Conference 2010 in Bodø. The NFB Conference is the annual conference of the National Research School in Business Economics and Administration; the acronym totally escapes me (Nasjonal Forskerskole i Bedriftsøkonomi?). Some of the plenary sessions looks interesting:

Thursday, September 2nd:
Time
11.00 – 11.45 Blockbuster Ideas – A really fascinating idea is the best way to success. How can we generate ideas and how do we know when they are interesting beyond our own little silo? Prof. Robert V. Kozinets, University of York, Canada A5
11.45 – 12.30 Research ethics: Pitfalls to be aware of for researchers and leaders. Prof Knut W. Ruyter, UiO A5
Friday, September 3rd:
Time
09.00 – 09.45 The Narrative Turn in Business Studies Prof. Larry Browning, University of Texas at Austin A5
09.45 – 10.30 Academic Writing: The craft of doing research Prof. Inger Johanne Pettersen, TØH