In this Saturday’s New York Times, in an article titled The Chasm Between Consumers and the Fed, I found the most amazing chart:
Of course I am not a financial journalist, so I have absolutely no understanding of the gigantic amoeba-like-shaded-area in this chart. But it looks very cool and very much like it represents something about which the article has much to say. Sadly, however, the New York Times does not provide the methodology it used in obtaining the amazing fact that six of the points can be grouped together while those other two points are excluded from the party. What astounding mathematical finance model did the Grey Lady use to come up with this plot (I’ll be it involves Ito calculus)?
Frustrated by the lack of transparency, I decided that it would be best if I tried to come up with my own methods and models for obtaining this graph. My first attempt, after scouring the economics literature and using some advance methods (related to integrating over Banach spaces) was the following
As you can see this model seems to pick out the overall rate of return as the defining characteristic. After much great reflection, and reacquainting myself with some obscure results from the theory of hyperbolic partial differential equations and new deep learning techniques from machine learning, I was able to tweak my model a bit and obtained the following
Now this is a beautiful plot, but it clearly does not reproduce the graph from the New York Times. What exactly, was I missing in order to obtain the giant amoeba of correlation?
But then I remembered…I’m not a financial journalist. I’m a physicist. And so, I took a look at the stats notes I took as a physics major at Caltech, quickly plugged in some numbers, and obtained a new, reality based, version of the plot
Well it’s not the New York Time plot. But I like it a lot.
Hoisted From the Comments: Funded Research
A while back Michael Nielsen
Re your last two paragraphs: a few years ago I wrote down a list of the ten papers I most admired in quantum computing. So far as I know, not a single one of them was funded, except in the broadest possible sense (e.g., undirected fellowship money, that kind of thing). Yet the great majority of work on quantum computing is funded projects, often well funded. My conclusion was that if you’re doing something fundable, then it’s probably not very interesting. (This applies less so to experimental work.)
This, of course, is quite a depressing idea: that the best work is funded at best indirectly by the powers that be. But it hadn’t occurred to me until much more recently that I, as someone who regularly applies for funding can do something about this problem: “My good ideas (all two of them)? Sorry Mr. Funding Agency, I’m not going to let you fund them!” And there is a bonus that if you submit something to an agency and they won’t fund it: well you can live under the illusion that you are doing might make the list of really important research.
Actually I’ve very proud of one research proposal I wrote that got rejected. The reviewers said “this work raises interesting questions” and then “but it’s just too crazy for us.” I mean it sucks to get rejected, but if you’re getting rejected because you’re just too crazy, well then at least you’re eccentric! (A similar story was my dream of becoming a ski bum after getting my Ph.D. in theoretical physics. I mean anyone can be a liftie, but a liftie with a degree in physics? Now that would set you apart! Lifties with Ph.D.s in physics please leave a note in the comment section of this blog 🙂 )
NRC CS Grad Ranking, Scientifically
NRC deranker. Scroll down to bottom for an explanation of the methodology.
"I sound my lonely trumpet in the dark trying to relax at the edge of precipice which once again faces me"
Steve sends me this gem, arXiv:0905.1039. The title of this blog post being a line from the paper:
Citation entropy and research impact estimation
Z.K. Silagadze
A new indicator, a real valued $s$-index, is suggested to characterize a quality and impact of the scientific research output. It is expected to be at least as useful as the notorious $h$-index, at the same time avoiding some its obvious drawbacks. However, surprisingly, the $h$-index is found to be quite a good indicator for majority of real-life citation data with their alleged Zipfian behaviour for which these drawbacks do not show up. The style of the paper was chosen deliberately somewhat frivolous to indicate that any attempt to characterize the scientific output of a researcher by just one number always has an element of a grotesque game in it and should not be taken too seriously. I hope this frivolous style will be perceived as a funny decoration only.
I wonder if this will provoke a response from my dear friend the Sad Physicist?
Not True in Any Base
Yes, dear Gray Lady, you certainly sound more sophisticated when you use the word “prime number” in your newspaper. But perhaps you might want to look up the actual meaning of the word before placing those words prominently beside two times five times five.
But I Thought…Click…Oh
Some confusing headlines:
Quantum Theory’s Release Date Now ‘TBA’
Quantum Theory Faces Delay Due to Quality Reasons
Quantum Theory Delayed
Mmmvelopes. Tasty Tasty Mmmvelopes.
Too often in life I am sending out a check to some charitable organization, or to resubscribe to Bacon magazine, and I think “damn this would be a lot better with Bacon.” And now via the honest one, I find out that there is a solution to this vexing problem: Bacon flavored envelopes! From the “learn more” section of the webstie:
Technology has given us a lot lately. The car. TV. X-rays. The refrigerator. The Internet. Heck, we even cured polio. But what have our envelopes tasted like for the last 4,000 years? Armpit, that’s what.
Really, people? If we can’t overcome this kind of minor technical challenge, it’s only a matter of time until some super-advanced race of aliens with lasers, spaceships and a delicious federal mail system comes down and colonizes the world. And nobody wants that (except for the aliens, of course).
So, after thousands of years and kajillions of horrible tasting envelopes licked, we’re happy to report that J&D’s Bacon-Flavored Mmmvelopes‚Ñ¢ are here to save the day. No longer will envelopes taste like the underside of your car. You can enjoy the taste of delicious bacon instead.
That’s right, bacon. It’s not real bacon, mind you, so you won’t have to start storing your envelopes in the refrigerator. But it really does taste like bacon. Which is what you really wanted in the first place, isn’t it? And it only took us 4,000 years to get there. Eat that, alien invaders.
Cool, but I beg to differ. My armpit smells like….Bacon!
Igon Value Problems Over Dilettante Matrices
Friday the 13th is, apparently, a day of must read articles. This time it’s Steven Pinker’s review of Malcolm Gladwell’s What the Dog Saw: And Other Adventures
He provides misleading definitions of “homology,” “saggital plane” and “power law” and quotes an expert speaking about an “igon value” (that’s eigenvalue, a basic concept in linear algebra). In the spirit of Gladwell, who likes to give portentous names to his aper√ßus, I will call this the Igon Value Problem: when a writer’s education on a topic consists in interviewing an expert, he is apt to offer generalizations that are banal, obtuse or flat wrong.
No Trail Email
From the annals of high idiocy, I enjoyed this sequence of emails at BofA:
“Unfortunately it’s screw the shareholders!!” Charles K. Gifford wrote to a fellow director in an e-mail exchange that took place during the call.
“No trail,” Thomas May, that director, reminded him, an apparent reference to the inadvisability of leaving an e-mail thread of their conversation.
…
Shortly after Mr. May’s remark about an e-mail trail, Mr. Gifford said his comments were made in “the context of a horrible economy!!! Will effect everyone.”
“Good comeback,” Mr. May replied.
You have to give Mr. Gifford at least credit for not replying back “OMG Oops!!!” after the first email exchange.
Helmet Heads Hijacked?
How did I miss this one from 2005? And how come no one told me to take off my tinfoil hat? Via @kmerritt, “On the Effectiveness of Aluminium Foil Helmets: An Empirical Study” by Ali Rahimi, Ben Recht, Jason Taylor, and Noah Vawter.
Among a fringe community of paranoids, aluminum helmets serve as the protective measure of choice against invasive radio signals. We investigate the efficacy of three aluminum helmet designs on a sample group of four individuals. Using a $250,000 network analyser, we find that although on average all helmets attenuate invasive radio frequencies in either directions (either emanating from an outside source, or emanating from the cranium of the subject), certain frequencies are in fact greatly amplified. These amplified frequencies coincide with radio bands reserved for government use according to the Federal Communication Commission (FCC). Statistical evidence suggests the use of helmets may in fact enhance the government’s invasive abilities. We speculate that the government may in fact have started the helmet craze for this reason.
Obviously