Getting it right and getting called out

Public mistrust of science: it is often in the public articles themselves. Here is an example: someone will notice that factor A is correlated with effect X. Journalist writes: “scientist show that A causes X” which, of course, isn’t the same thing. In fact, many (most?) correlations have nothing to do with causation, but they give a place to look.

Many who write articles often don’t understand things like:
1. False positives. Lots of times, especially in the biological sciences, a statistically significant effect has up to a 5 percent chance of being a false positive.
2. Sometimes scientists themselves overplay the significance of their results.
3. Sometimes, the result is a very narrow one. Example: a training technique might have been tested on, say, university level runners and show a 1 percent improvement on their time for, say, a 5K run. But that effect might only apply to university runners (e. g. runners of that age, ability and experience).

Or it might be something like this: a test might show that those who drink 2-3 cups of coffee today enjoy a 5 percent reduction on a risk factor for a given disease…and this result gets a “coffee is good for you” headline. Another study might show that 3 or more cups of coffee a day might increase a risk for another unrelated disease and lead to a headline: “study shows that coffee is bad for you.”
4. Innumeracy. Yes, drug X which treats a condition might double the chances of getting a certain fatal disease! That is terrible, right? Well, maybe not if the risk of that fatal disease goes from 1 in 10,000,000 to 2 in 10,000,000. But maybe so if the risk goes from 1 in 10 to 2 in 10. Context and data matter!

Jon Stewart: here is his segment on the NRA convention. The best part, IMHO:

1. the conservatives accuse us of “using fear”. So what do you think that they do?
2. Senator Ted Cruz: brags about filibustering. He then accuses Democrats of “tyranny” when they filibuster.

Paul Krugman
He points out that Niall Ferguson has really been whiny (he is the one that tried to discredit Keynes by pointing out that he was gay). He then directs us to a William Black article that just takes Ferguson apart:

We now have Niall Ferguson, a history professor at Harvard and Hoover fellow whose theoclassical views have proven so influential with Prime Minister Cameron’s government’s adoption of austerity policies that have killed the UK recovery.

Ferguson has had a terrible last 10 years. He was a strong proponent of invading Iraq and pines for us to stay indefinitely in Afghanistan. He was a Romney supporter who wrote an anti-Obama screed in Newsweek that demonstrated his contempt for facts. He was an advisor to Senator McCain’s campaign for the presidency. Because of his record of getting every important policy issue wrong he was paid a great deal of money to speak to an “alternative investment” conference that began, with no small irony, on May Day. Ferguson was presenting his thesis that the West has become “degenerate.” He certainly proved that point about himself.

Krugman v. Ferguson (2009) (TKO for Krugman in the First Round)

Ferguson has been spinning out of control in recent weeks. In 2009, he made the mistake of trying to debate a Nobel Laureate in Economics (Paul Krugman) about Krugman’s specialty. If it had been a fight the ref would have stopped it in the first round and awarded a TKO. Ferguson did his ode to austerity as a response to the Great Recession and claimed that the stimulus program was causing, and would continue to cause, interest rates to soar and prevent a recovery.

Austerity created the gratuitous über-Depression in the Eurozone’s periphery. U.S. interest rates have fallen to record lows. Ferguson admitted recently that stimulus had not produced his predicted surge in interest rates and that austerity in response to the Great Recession proved self-destructive. That was fine, but Ferguson could not leave it there. He added three points got him trouble – and those three points prompted Ferguson’s latest and greatest of own goals. First, Ferguson tried to reinvent the history of the position he took during the 2009 debates with Krugman.

Second, having agreed that Keynes had proven correct and Ferguson had again been proven incorrect in his predictions, Ferguson proceeded to continue to demonize Keynes as the cause of much of the West’s supposed degeneration. This is more than passing strange because it is Ferguson’s policies that have proved disastrous and Keynes’ policies that have proven correct.

Third, Ferguson responded furiously on March 6, 2013 to Krugman’s article pointing out Ferguson’s effort to air brush out of history Ferguson’s history of predictive failure. Ferguson’s cri de cœur is so delectable because it sends a frisson through one’s body to see such naked hypocrisy and whining in print from the self-proclaimed champion of American military adventure designed to create and expand a new American empire and a writer whose works are now redolent of innuendo.

“In my view Paul Krugman has done fundamental damage to the quality of public discourse on economics. He can be forgiven for being wrong, as he frequently is–though he never admits it. He can be forgiven for relentlessly and monotonously politicizing every issue. What is unforgivable is the total absence of civility that characterizes his writing. His inability to debate a question without insulting his opponent suggests some kind of deep insecurity perhaps the result of a childhood trauma. It is a pity that a once talented scholar should demean himself in this way.”

Ferguson, trained as a historian, uses innuendo to make up a (self) “suggest[ed]” history to smear a critic who (1) proved correct, (2) proved Ferguson incorrect, and (3) correctly called out Ferguson’s effort to change history to mislead readers about point #2. But what sends the frisson through your body when you read Ferguson’s effort to smear Krugman is Ferguson’s naked hypocrisy – and his blindness to it. As with his repeated efforts to smear Keynes, Ferguson’s attempt to smear Krugman reveals everything important and true about Ferguson and nothing important or true about Keynes or Krugman. The nice thing about Ferguson is that despite the adage that “practice makes perfect” he is getting ever cruder and more self-destructive in his efforts to smear those with whom he disagrees.

Black goes on to gut Ferguson….read the rest.

But back to what Krugman was writing about: he gives 5 other instances in which a conservative critique of Keynesian economics failed miserably. He finishes with an open mocking of Ferguson:

So, if I were Ferguson I guess I’d have to seek some kind of psychosexual explanation here. I would note that none of these guys has a beard. Masculinity issues?

A bit of math in the “real world”

Paul Krugman

I see that Simon Wren-Lewis is on a campaign against the use of the Mundell-Fleming model, a simple international macroeconomic extension of IS-LM analysis, because of the way it handles the relationship between interest rates and exchange rates. And it’s true that the simplest version of Mundell-Fleming assumes that interest rates are equalized by capital flows, taking no account of expectations of future exchange rate changes.

But is that the way it’s taught? It’s certainly not the way the issue is handled in the leading undergraduate textbook in international economics, which for nine editions — 25 years! — has worked with an exchange rate model in which investors expect the exchange rate to revert to a long-run norm, so that you get a downward-sloping relationship between the interest rate and the price of foreign currency:

Business calculus question: if then what is the sign of ? What about the sign of ?

Genetics and Evolution
Larry Moran on the Hardy-Weinberg Theorem for genetic evolution:

Imagine that you have a population with two alleles, A and a, at a single locus. The frequency of the first allele is f(A) to which we assign the value p. The frequency of the second allele is f(a)=q. In a randomly mating sexual population the probability of an A sperm being produced is p and the probability of an a sperm being produced is q. Similarly, the probability of an A egg cell is p and the probability of an a egg cell is q. These probabilities, p and q, do not have to be equal.

We can calculate the probabilities of all possible combinations or sperm and eggs in the population from a the following diagram (Punnett square).

The idea is that you calculate the probability of AA, Aa, aA, aa. Not surprisingly, , which of course, means which follows from . No biggie. What this means is that for an ideal population (large enough for which the probabilities are realized, the allele frequencies do not change, ABSENT some other factor…..such….as…well, selection pressures, migration (populations leaving the area), mutations (new alleles) or drift. When the equilibrium starts to fail, evolution has taken place.

But…many ID types don’t know that.

science topics: jet stream and weather, support for gay marriage, ID, isolation, evolution oddities, etc.

Workout notes
Weights only:
rotator cuff
pull ups (5 sets of 10); hip hikes and Achilles exercises
incline press: 10 x 140, 4 x 155, 6 x 150, 7 x 145
abs: 3 sets each of crunch (10), twist (10), sit backs (10), vertical crunch (20)
dumbbell military: 3 sets of 12 x 50
rows: 3 sets of 10 x 65 (each arm)
dumbbell bench: 2 sets of 10 x 65
curls: 2 sets of 10 x 57.5 pulley, 1 set of 10 x 30 dumbbell
pull downs: 3 sets of 10 x 160
I’ll run 4 times a week until I can get over this foot soreness.

Posts
For those interested in physics, physicist Mano Singham has a multi-part series on the Higgs Boson; he is giving you some background for it:

Part: I, II, III, IV, V

Social
There is evidence that being socially isolated harms longevity, even in the absence of feelings of loneliness.

Opinions on same sex marriage: favorability IS going up across the board; no doubt about that. But for a detailed analysis: read Nate Silver’s post:

It is also possible to project how the results in each state might change over time. I assume that support for same-sex marriage will continue to increase by one and a half percentage points nationally per year, which reflects the recent historical trend from both polling and ballot-initiative data. (The way that the model is designed, support might be projected to increase slightly faster or slower than that in individual states based on the number of swing voters.) Thus, we can extrapolate the results forward from 2008 to 2012, and to future years like 2016 and 2020.

Roughly speaking, by 2020, only a few states in the deep south will have less than a majority favoring same sex marriage. The median state support (median of all states) will be about 60 percent. Time is marching on and I hope that Illinois stays ahead of the curve.

Climate
What is driving our crazy weather? Conjecture: we have a steep increase in sea ice, which leads to the water having more heat, which leads to a change in the path of the jet stream, which allows that cold arctic air mass to dip down lower than before (in sort of a sine wave type path).

Biology
How about a fish with a transparent head?
Read more at the link; note those green things are…they eyes! The eyes are INSIDE the head!

Read more at Jerry Coyne’s website.

Evolution deniers
If you are going to try to deny established science, it helps to know what you are talking about. ID types, in general, don’t. No, information theory does NOT disprove evolution. And no, Larry Moran is NOT a creationist, though he ascribes a bigger role to genetic drift and a lesser role to natural selection than, say, Jerry Coyne does. But that is a scientific dispute on mechanisms of evolution, NOT a questioning of whether evolution took place or not. This is (sort of) analogous to the various interpretations of quantum mechanics:

But none of these deny quantum mechanics.

Home: getting caught up.

Yep, that is me, at a rest stop between Cleveland and Chicago. I dressed up for my daughter’s university visit.

Mathematics Mano Singham has an interesting post on the “transitive property”; for real numbers this is and implies . For real numbers (or other totally ordered sets), this is no big deal. But there are some situations in which objects that would seem to have such an ordering really don’t.

Here is such an example. Consider dice labeled as follows:

Red beats green 4 out of 6 times. Green beats blue 4 out of 6 times. Blue beats red 4 out of 6 times. All the colored dice tie the “normal” die. More here.

Politics
Yes, it is great that a Republican Senator came out in favor of gay marriage, now that he found out that he has a gay child. Well, if it is really the right thing to do: why not come out prior to this?

Social
here is a list of “sad states”:

A new well-being ranking shows the United States has an alarming ‘sadness belt’ in the South and Midwest where many residents are depressed, obese and disgruntled at work.
Kentucky, Mississippi and Tennessee scored almost as low as West Virginia in the annual Gallup poll, which looks at six categories including life evaluation, emotional health, work environment, physical health, healthy behaviors and basic access.

What I found interesting: so many of the cold weather states are on the “happy” side. Why????
I suppose it is because those who live there WANT to be there.

Social

Interesting. Yes, someone else’s obesity isn’t my business. And yes, obesity doesn’t measure pleasantness, intelligence, etc. And yes, too often, fat people are dismissed as being competent merely because of their size.

But this photo is clearly about physical attractiveness. And guess what: we find attractive what we find attractive, period. No one tells me “you had better find this person to be sexually attractive”; I have a built in mechanism that does that for me.

Note: while I often joke about women runners in spandex, I tend to find those who run the 5K in the 25-33 minute range to be the most attractive; they are fast enough to have some fitness but not so fast that they are all skin-and-bone. Though, yes, some of the faster ones are pretty cute too; these days women at those faster speeds tend to be too young for me. Not ALL of them though.

Another example of projections (statistical inferences, mathematical maps) ….

I wrote about the loss of information that comes with projecting something onto a “smaller” set of numbers (or onto fewer dimensions) here.

Here is roughly what I am talking about:

Imagine seeing a shadow of a person on the ground. Could you imagine two different people casting (almost) exactly the same shadow? Basically, when you project 3 dimensional information onto a 2 dimensional surface, you lose information. Here is a spectacular example of that:

I was reacquainted with this concept today.

1. Last night, some of my friends went running during a sleet storm. It was 32-34 F and windy. This morning: I want running when it was 22-24 and windy (somewhat less so, but still windy)…but my run was more pleasant than theirs even with a 10 degree colder windchill. Why? Easy: no sleet; no icy stuff blowing in my face. Projecting conditions onto two numbers (temperature and wind speed) loses information.

2. Biographies. Imagine this: suppose you wrote a story about yours truly. You focused on the following true events:
1. I finishing low in my class at nuclear power school.
2. I finished my freshman year poorly both militarily and academically.
3. I dnf’ed several long races and performed poorly in some 12/24 hour ones.
4. I had papers rejected for publication; a couple had embarrassing errors in them and some had some unflattering comments from the referee/editor.
5. I failed the topology preliminary examination twice in graduate school, had a rough first year, and failed my first engineering officer of the watch board while on the submarine.
6. I was morbidly obese at one time in my life.
7. I’ve been divorced.

All of the above are true and from those events, you’d gather that I did NOTHING right at all (some might say that was a fair conclusion! )

But, by ignoring (or soft-peddling) the above and switching the focus to the things I did right and to my successes (yes, I’ve had some…let me think….I’ll get back to you…;-) ) well, you’d get a different picture of me.

So, my point: biographers can do the same to public figures; one can remain 100 percent truthful and paint very different portraits of a public figure. This Slate article talks about Bob Woodward and the biography that he did of John Belushi.

Note: Woodward came under more scrutiny when we complained that some Obama official said that “he would regret” passing along false information. You can see the e-mail messages here; the right wing and other Obama detractors attempted to use this as “evidence” of the belligerence of the Obama administration.

Weekend coming up

Workout notes
Weights. I was stronger than expected:
rotator cuff
pull ups (5 sets of 10)
bench: 10 x 135, 5 x 180 (got a spotter), 6 x 170, 10 x 160
incline: 9 x 135, 9 x 135
pull down/curl/Hammer row supersets: 3 sets of 10 each (160 pull down, 210 row, 57.5 curls; more than normal)
military (dumbbell): 2 sets of 12 x 50, 1 set of 10 x 80 machine.
abs: 3 sets of: crunch, twist, sit back, vertical crunch
legs: 3 sets of: adduction, abduction, push back (slowly done)
Also I did 100 hip hikes (each side), back PT, and more piriformis/gluteus medius.

Sadly, the leg exercises were enough to induce some mild “after the fact” quivering; I am weak.

This weekend: I am singed up for a trail marathon. I have to tell myself: unless I am injured, quitting isn’t an option. That also means going out very conservatively and bringing extra shoes, gaiters and plenty of salt tablets.
The trails might be a mess, so perhaps 3:15/3:45 = 7 hours is realistic. If I am not injured and the RD hasn’t pulled me from the course, I go onward after the first loop.

Posts
John Roberts is known as a brainy justice, but evidently he cherry picked some statistics during his Voting Rights Act examination. He also might not have used the statistics correctly.

This brings me to another difficulty that I have: I am reasonably well versed in statistics and some scientific matters. But what about, say, macro economics? I have no economic training.

I ask that because I follow Paul Krugman and I like his stuff. But do I like his stuff because it agrees with my (undeveloped) intuition and ideology? I honestly can’t tell and I too can fall victim to believing “what I want to be true”.

Who are the bad guys? Who are the good guys?

First, a bit of mathematics and physics in action: coffee rings and randomness. There are several ideas in this “non-technical but non-insulting” article. One is that the shape of the tiny particles themselves help shape the patterns that they eventually form. Another: there is a nice lesson in randomness here. Please surf to the article to see an animation of a random Poisson process in action.

Conservatives vs. Liberals
On a personal level, conservatives (on the whole) are not that bad. In fact, it appears that they will intervene more quickly to help someone out:

And some research indicates that they might be more generous with individual charity, volunteering and blood giving. Yes, I know: blood donation rules are biased against liberals (no gay men, no one who has spent too much time abroad in certain countries (like the UK!)) and I know that much of conservative individual charity goes toward sprawling churches (of course, liberals tend to give to museums…yep..guilty…which is hardly “alms for the poor”). On the whole, I think that conservatives ARE better in this area.

But when it comes to the POLITICAL LEADERSHIP (and no, Democrats can’t point a finger about corruption: here, here), well, I think that too many Republicans are both exclusionary and too bent on short term political gain.

As far as being exclusionary: even Newt Gingrich sees it.

As far as being too focused on winning a short term political battle: the Republicans simply won’t negotiate. This comes as no surprise to many of us.

And as far as their “policy experts”: much of the time they consist of snake oil salesmen and credentialed people of limited competence:

In the article, the case for slow growth forever is mainly made by quoting Kevin Warsh, a former Fed governor. And Warsh is indeed someone who has been wrong about everything; a bubble denier who spoke of strong capital markets before the crash, a hawk who has been warning about the risk of inflation for three years, an invoker of invisible bond vigilantes who somehow managed to describe the supposed threat from these vigilantes as somehow both a certainty and unknowable.

If there is a special distinction to those of Warsh’s speeches and articles I’ve read, it’s this: he has had a habit of saying and writing things that are supposed to be profound, but say nothing at all. Can anyone tell me the point, if any, of this WSJ op-ed?

But wait: who is Kevin Warsh, anyway? Well, he’s a lawyer turned investment banker turned Bush appointee to the Fed turned Hoover fellow — not an economist at all. Now, I hate credentialism: there are plenty of fools with Ph.D.s, some fools with fancy prizes, and a fair number of first-rate economic thinkers without formal qualifications. Still, if someone is going to make pronouncements about how the whole nature of the business cycle has changed, you’d like some sign that somewhere in his life he has thought hard about, well, anything.

So why pay any attention at all to this guy on these matters? I guess it’s a different kind of credentialism — the Beltway notion that because somebody was once appointed to a policy position, he must be an expert. But that is, of course, ridiculous — and people at the Washington Post, who get to see former officials all the time, surely must know better.

Thanks Paul Krugman!

(cons vs. libs charity)

http://whyevolutionistrue.wordpress.com/2013/03/03/in-which-i-once-again-graylings-bulldog-biting-peter-hitchens/

Drones, guns, bacteria, Republicans…and evolution

Politics
This is a link to a 16 minute Maddow show segment on the current infighting among political conservatives. Roughly speaking: you are seeing a rift between the rabid populists and the wealthy plutocrats. Though I am loving this, I remember that the Democrats also looked in disarray in 2005; it did NOT look pretty.

World Events
Yes, drone warfare sucks. But other kinds of warfare: sucks worse. The civilian casualty rate is higher with conventional bombing and military actions. (article by William Saletan)

Statistics in the News
Some gun advocates said this:

In 1966 there were 36 rapes per 100,000 people in
Orlando, triple the 1965 rate. In 1967, there were 4.
Before the [firearms training for women program] training, rape rates had been increasing in
Orlando as nationwide. 5 years after the training, rape
was still below pre-training levels in Orlando, but up
308% in the surrounding areas, 96% for Florida overall,
and 64% nationally.

So, is this convincing? Well…not really (via Tim Lambert):

To do an honest comparison we need to do look at more than one year.

58-66 67-72
Orlando 15.5 19.1
Surrounds 12.2 23.9
The rate did not increase by as much as in the surrounding area, but
this is not good evidence for a long term benefit.

What about a short term benefit? Kleck [1] admits that the rate was
extremely variable, but claims that because the change exceeded two
standard deviations, random variation was insufficient to explain the
change. However, in 64 there was a change almost as large and also
exceeding two standard deviations.

Kleck also states that the percentage decrease was larger than in any
other US city with a population of over 100,000. Kleck neglects to
tell us what the population of Orlando was, but by looking at the
granularity of the data you can deduce that the population of Orlando
was less than 100,000 for the whole period 1958-1972. Comparing
apples with oranges. Cute, real cute. Orlando itself experienced a
larger percentage decrease in 1963.

To summarize, the variability in the data is sufficient to explain the
change, [...]

Moral: trying to extract something from a momentary change in very noisy data is dangerous.

Note: during that year, homicide rate increased by 22 percent!!! Oh well…talk about “cherry picking” your metric.

Medical Science
This is a nice article about medical tests and procedures: many are unnecessary. Here is a good example from my life: in 1999 I had been having chest pains. Mostly they came when I was in the prone position (lying down) and they got better when I got up and walked around; I had no problems when I ran. When I saw the doctor, he told me: “I could stress test you but the chances of a false positive are so high, we’d learn nothing from the test. I could do an upper GI series on you, but we’d end up just treating the symptoms anyway, so why don’t we do that?” So I got prescription strength antacids and sure enough, it went away…and never came back.

Evolution and bacteria
Experiments on e-coli bacteria have shown that, under stress, organisms (at least bacteria) evolve in a somewhat predictable way. That is, different cultures under the same stresses evolved very similar solutions, often by the same mutations. Of course, this organism reproduces asexually and this drives down potential variations.

References: here, here, here, here and here.

Some math in “Better Angels of our Nature” by Pinker

I am reading The Better Angels of our Nature by Steven Pinker. Right now I am a little over 200 pages into this 700 page book; it is very interesting. The idea: Pinker is arguing that humans, over time, are becoming less violent. One interesting fact: right now, a random human is less likely to die violently than ever before. Yes, the last century saw astonishing genocides and two world wars. But: when one takes into account how many people there are in the world (2.5 billion in 1950, 6 billion right now) World War II, as horrific as it was, only ranks 9′th on the list of deaths due to deliberate human acts (genocides, wars, etc.) in terms of “percentage of the existing population killed in the event”. (here is Matthew White’s site)

But I have a ways to go in the book…but it is one I am eager to keep reading.

The purpose of this post is to talk about a bit of probability theory that occurs in the early part of the book. I’ll introduce it this way:

Suppose I select a 28 day period. On each day, say starting with Monday of the first week, I roll a fair die one time. I note when a “1″ is rolled. Suppose my first “1″ occurs Wednesday of the first week. Then answer this: “what is the most likely day that I obtain my NEXT “1″, or all days equally likely?”

Yes, it is true that on any given day, the probability of rolling a “1″ is 1/6. But remember my question: “what day is most likely for the NEXT one?” If you have had some probability, the distribution you want to use is the geometric distribution, starting on Thursday of the next week.

So you can see, the mostly likely day for the next “1″ is Thursday! Well, why not, say, Friday? Well, if Friday is the next 1, then this means that you got “any number but 1″ on Thursday followed by a “1″ on Friday, and the probability of that is . The probability of the next one being Saturday is and so on.

The point: if one is studying the distribution of events that have a Poisson distribution (probability ) on a given time period, the overall distribution of such events is likely to show up “clumped” rather than evenly spaced. For an example of this happening in sports, check this out.

Anyway, Pinker applies this principle to the outbreak of wars, mass killings and the like.

Enigma Machines: how they worked, and the flaw…

Note: this type of cipher is really an element of the group , the symmetric group on 26 letters. Never allowing a letter to go to itself reduced the possibilites to products of cycles that covered all of the letters.