"... recall the famous quip from Jeff Hammerbacher, a former Facebook engineer, who said, 'The best minds of my generation are thinking about how to make people click ads. That sucks.' We contend that working in biostatistics using complex neuroimaging data to solve the biggest puzzle of them all – understanding the human brain – is a worthy alternative."-from "WANTED: Neuro-quants" by Brian Caffo, Martin Lindquist, and Ciprian Crainiceanu, which appears in the the Huffington Post.
Showing posts with label statistics. Show all posts
Showing posts with label statistics. Show all posts
August 14, 2013
Proposed: Understanding the brain doesn't suck.
August 6, 2013
July 18, 2013
Title of the Week!
"Strobing Brain Thunders: Functional Correlation of Extreme Activity Events" by F. Lombardi, D.R. Chialvo, H.J. Herrmann, and L. de Arcangelis, appears in the journal "Chaos, Solitons & Fractals."
June 24, 2013
Your functional neuroimaging update (with study questions).
1. "Graphs are the new blobs."
Discuss. How are graphs and blobs similar? How are they different?
2. A leading search engine gives no hits on that phrase:
Do you expect this to remain true for long?
May 31, 2013
"Bucky is not like other dogs. But he is like ..." Part III. ("Visually similar images" from Google's "search by image")
 |
| (click on an image for full-screen) |
May 29, 2013
February 21, 2013
Compelling and Devastating.
"The Extraordinary Science of Addictive Junk Food" by Michael Moss will appear in this Sunday's New York Times Magazine.
December 6, 2012
Are you a methodologist or an ASS?
In the article "Primary study authors of significant studies are more likely to believe that a strong association exists in a heterogeneous meta-analysis compared with methodologists" in the Journal of Clinical Epidemiology, Orestis A. Panagiotou and John P.A. Ioannidis report that "authors of significant studies" (a group indicated by the acronym "ASS" in the paper's Tables; for example Table 2 has a column labeled "ASS vs. methodologists") are "influenced by their own investment in the field" and so are "more likely to believe that a strong association exists."
November 4, 2012
Is abuse of the null hypothesis a felony or a misdemeanor?
Just one sentence from "Activity Metabolism Measurements Made Easy in Metabolic Phenotyping" by Dr. J.R.B. Lighton:
"Thus the null hypothesis is not only falsified, it is smashed into smithereens that daintily and prettily drift down to the floor like dandelion seeds, except that they sparkle in the moonlight."
h/t: Dr. M.R. Laughlin
October 31, 2012
October 12, 2012
Correlation and um, y'know, the other thing...
"Chocolate Consumption, Cognitive Function, and Nobel Laureates" appears in the October 10th issue of the New England Journal of Medicine.
May 17, 2012
"Finally, SUPEREGO-related neural activity was localized to lateral prefrontal and parietal cortex."
"A Triple Dissociation of Neural Systems Supporting ID, EGO, and SUPEREGO" (because "... Social Neuroscience, Neuroeconomics, and Developmental Social Cognitive Affective Clinical Neuroscience are just not cutting edge enough anymore...") appears in Psyence.
e.g., "Rule number eight: exploit superstitious thinking"
"Ten ironic rules for non-statistical reviewers" ("... written for reviewers who may not have sufficient statistical expertise to provide an informed critique during the peer-reviewed process, but would like to recommend rejection on the basis of inappropriate or invalid statistical analysis") by Prof. Karl Friston will appear in the journal NeuroImage.
April 27, 2012
On average, each bike has a fork.
On the other hand (h/t: cycleexif.com), Olli Erkkilä's beautiful and poetic Graduation Project at the Institute of Design in Lahti, Finland was a "forkless" cruiser:
February 11, 2012
April 13, 2011
A modest proposal...
Now that it has been reported that "... greater liberalism [is] associated with increased gray matter volume in the anterior cingulate cortex, whereas greater conservatism [is] associated with increased volume of the right amygdala", we have no doubt that candidates for high political office in the USA will soon be expected to disclose not just their tax returns and medical evaluations, but also their brain scans.
February 22, 2011
A perverse incentive?
"Publish or perish" is so twentieth-century. Today, it is not enough to publish – your papers must be cited. Citation indices abound, and many a scientist monitors their own h-index.
So, as citations have become such an important metric, a shrewd strategy for professional advancement would be to publish papers that are wrong.
Because then, others will publish papers correcting yours, and cite you in the process, thereby making you a highly-cited researcher.
January 28, 2011
"A man's got to know his limitations."
As you may have heard, on Wednesday we got some extraordinarily heavy wet snow here in the National Capital Area. Only about half a foot of snow, mind you, but much much heavier than any half-foot snow-fall ought to be, because it was really sopping wet. As a result, our area suffered from extraordinary levels of tree damage and power outages (because so much electricity here runs in overhead power lines).
At home, one of our street trees dropped a big limb, blocking the street in front of our house. So, I went out – wearing lug-soled boots & winter layers – to try to move the limb. I determined that it was far too heavy for me to move.
What about snow plows? Would the next plow to come along just push the limb out of the way? Hard to tell – to me, the limb was neither clearly so small that it could easily be moved by a plow, nor clearly so large that it could not.
What would the plow driver decide? Might different plow drivers make different decisions? Well then, a probability distribution would be required to compute the likelihood of the street remaining blocked over time.
I was wondering how big an error would be made by using a Gaussian, as opposed to a Poisson distribution, when a van drove up, and stopped. Two men wearing sneakers, sweatshirts, and jeans got out, moved the fallen limb aside, and got back in. The van drove through, and away.
I went out with a rake, to clear the many smaller broken branches from the snowy street, and I thought: Some problems are best solved by persons who are not scientists.
___
January 6, 2011
January 2, 2011
Inversion as a good.
In the 1940's, Jerry Cornfield & Wassily Leontief invented econometrics. In one of their first studies on the "sectors" of the American economy, they used economic data to arrive at a 24 x 24 matrix that they then needed to invert. Inverting that 24 x 24 matrix by hand, would, they calculated, take several centuries.
Harvard University's IBM "Mark I" electro-mechanical computer, developed for the war effort, was in principle available to perform this calculation, but the work would have to be paid for, and Cornfield & Leontif's bosses at the Bureau of Labor Statistics refused to approve such an expenditure, because, according to policy, the government would pay for goods, but not for services – after all, so many experts were on the government payroll that they saw no need to pay for services.
In the end, the bureau issued a purchase order, for "one matrix, inverted" from Harvard.
Subscribe to:
Posts (Atom)
