Giving Thanks

Al Gore and David Blood have written an essay:

"the basic building blocks of commerce and markets: accounting, disclosure, incentives, regulation and responsibility."

This is an important Economic point that many people don't understand. Are supply and demand for products and services the basic building blocks of commerce and markets? No. The basic building blocks are accounting and regulation.

The farmer down the road knows how to produce corn. I want to eat corn. Is that what leads to the creation of a farmers' market? No, that is a common misconception. What creates that vegetable stand are things like government taxes and agriculture subsidies. Like paying the farmer not to plant corn, for example. And accounting, of course. Farmers benefit from accounting and filling out paperwork for government bureaucrats because it helps bring in the harvest. This Thanksgiving, let us all give thanks for the government incentives, regulations and disclosure requirements that put food on our table and make our lives worth living. Bless you Al!

I thought turkeys could fly!

If it's Thanksgiving, it's time for the WKRP Turkey Drop

Prepped and ready

Things are almost ready for tomorrow. The ham was picked up on Monday morning. Tuesday was for making the cranberry bread. Today, the turkeys (2!) got picked up, 2 pumpkin pies were made, cranberry sauce is ready, rosemary fingerling potatoes just need heating up, the stuffing is cooked and ready for reheating, and finally, both turkeys are sitting in brine for the night.

Tomorrow, we need to make the gravy, get one turkey in the oven and the other on the rotisserie grill, reheat everything, possibly make wild rice, and we're done!

All of that food (including Bahamian beans and rice and a pie that others are bringing), and we are only going to have 7 adults and 4 kids! Yikes!

I think we'll have leftovers until 3 years from now.


According to a recent report, Massachusetts is ranked 3rd in obesity, while Tennessee is ranked 49th among the states.

So is it noticeable to a person who lived in Massachusetts for many years and recently moved to Tennessee?

You bet. There are occasional times here when I look around and am taken aback by the number of people who are obese. I don't mean a bit of middle-age paunch, I mean really morbidly obese. I believe the report, and believe that it is a public health problem for Tennessee.

What can be done about it? I'm not confident that government awareness programs are going to help much. One thing that may help is increased levels of income. In the USA, the poorer people are the fatter they are. I predict that Tennessee's obesity problem will get better with time as living standards rise. Until then though, I don't expect much change.

Shorewood Football

This is pretty weird. At the high school from which I graduated:

At virtually every high school, the traditional recipe for homecoming is a Friday night, home football game and packed bleachers.

Not Shorewood.

Instead of varsity players battling it out on the football field, there was a powder-puff game between junior and senior girls that ended in a tie and a dance-off at the 50-yard line.

Football is dead. Killed off by soccer! Who'da thunk it.

Ann says: a couple years after I graduated, the school shut down its football program (probably in '87 or '88). I think that only lasted a year or two before they put it together again. For at least one year the "homecoming" game that Homecoming referred to was that of the boy's soccer team.

TKO for AGW! (III)

Here are a series of e-mails that show how this work was being done. They all come from a single e-mail written 10 years ago--has anything changed since?: # 926947295.txt. Since these e-mails have several layers of replies, the earliest messages start at the bottom in the link. I will unravel them here and remove the ">" forwarding marks and clean up some paragraphing. The only thing I've changed is the removal of some phone numbers. All bolding is mine:

Here's the start of the e-mails, which is a correction to a previously-sent e-mail:
Mike MacCracken [Ann - this might be him ]
Date: Friday, May 14, 1999 8:12 PM
Subject: Re: CO2

Please disregard the previous message and replace with this message (1st
paragraph is unchanged).

On Fri, 14 May 1999, Benjamin Felzer wrote: [Ann - This might be him ]

> Going back to some of the original radiative forcing values, it would
> appear that the 1% increase is true of RADIATIVE FORCING, whether of CO2,
> CH4, etc, or the total (to an approximation). However, once we convert
> back to CO2 concentration (using the exponential relationship), the
> actual
> increase in concentration is more along the order of 0.7% (all
> compounded). Is it possible that the original 1% assumption was
> mistakenly applied to CO2 concentrations for the modelers when it was
> actually meant for radiative forcing??
> Therefore for the ecological models we should use Dave's original
> suggestion, because the models really did use a 1% increase in equivalent
> CO2, which approximates a 1% increase in CO2 alone. The point here is
> that this 1% increase is much higher than IS92a, but that might be because
> of the confusion between radiative forcing increase and concentration
> increase discussed above. In fact a 0.7% increase in equivalent CO2 might
> have been a more realistic assumption for IS92a, but the 1% increase in
> concentration is what was actually used in these earlier models. The CO2
> concentrations used in the ecological model should correspond to those
> used in the GCMs, not to what we think they should be.
> Any other thoughts?
> Ben

And now the reply to that one. I've tried to clean up the paragraphing:

-----Original Message-----
From: franci []
Sent: Saturday, May 15, 1999 3:58 PM
To: Benjamin Felzer
Cc: Mike Hulme [Ann - This is probably him ];;;;;; Mike MacCracken
Subject: Re: CO2

dear ben,

You just showed that the Hadley transient run we are supposed to use for the national assessment is too high, forcing-wise, because it assumes an overall 1.2% increase in total forcing.

My question is then the following:

-why are we using a 1% annual increase in GHG [Ann—I assume this is Green House Gas] forcing (corresponding to the 1.2% increase) as a criteria for GCM simulations to then be used for the national assessment? Is it because of the possible confusion you refer to below? If so, that criteria needs to be revised.

I still have a problem with the real CO2 calculations, in connections to hadley or CCCM. It seems to me it is still arbitrary to use one or another CO2 curve.

However, in this arbitrariness, two easy solutions are possible ( i am just summarizing previous e-mails, at the cost of being highly repetitive and obvious):

-one is dave's, i.e, assume no change i GHG forcing mix from today, and apply 1% compounded increase to the 1990 actual levels. That gives a concentration of real CO2 in 2100 that is >1050 ppm. THAT'S 50% higher than projected by IS92a, and even 17 % higher than the worst emission case devised in IS92f.

-the second is tom's. Just use the co2 in IS92a, and assume that all other further changes necessary to get the hadley forcing (whatever they are) happen in GHG other than CO2.

I will repeat that I like the latter solution.

Whatever the consideration of self-consistency and physics are when you make this decision, I do not think we should carry out the national assessment by using "unrealistic" CO2 numbers. I thought the numbers that come out of our exercises (from the impact side of things) were supposed to serve as some basis to be used in the process of decision making at the national and regional level. Am i out of line here? There are dozens of people right now, out there, including our group at giss, who are gathering data, fine-tuning models, making connections among physical and socio-economic variables, etc., at a very low "effort spent/retribution received", and then we are going to run things at 1000 ppm in 2100? As far as my specific contribution is concerned, it surely might make a big difference in crop yield changes under climate change whether I use 700 ppm in 2100 (the IS92a) curve, or >1000 ppm (the 1% compounded increase).

The problem is the same for the 2040's (the other decade we have decided to simulate), although possibly not as bad as the 2090's case.

Either solution we opt for, we have to make clear to whomever will receive our results that the climate forcing scenario is on the "high" side of things.

Ah! It was so nice and easy when we were working with doubled-CO2 equilibrium runs!



PS what about the CCCM scenario?

1) "supposed to use" according to who? or according to what scientific evidence? In general, this isn't the way science is supposed to happen. You do literature searches to back up your choices. If you choose one method of analysis or one value over another, you need to back up that decision with previous research. You aren't supposed to pull numbers out of the ether just because those are the ones you are "supposed to use". If you're a scientist, back up your choices with proof!

2) "It seems to me it is still arbitrary to use one or another CO2 curve." No kidding. And isn't this the crux of the problem? An arbitrariness in the selection of data and analysis techniques. In other sciences, this isn't a question. You can make an arbitrary decision, as long as you back up the reasoning behind your selection. Whether or not that was done is the heart of this whole matter.

3) "Just use the co2 in IS92a, and assume that all other further changes necessary to get the hadley forcing (whatever they are) happen in GHG other than CO2." Just pick something (whatever it is), make lots of unsubstantiated assumptions, and enjoy!

4) "Whatever the consideration of self-consistency and physics are when you make this decision, I do not think we should carry out the national assessment by using "unrealistic" CO2 numbers." What a novel idea! Our assessments should be based on realistic data! Shocking! Astonishing! What a breakthrough in scientific reasoning!

Next e-mail in the chain (I snipped out the telephone numbers on this one, no need to pass those along.):
On Sun, 16
May 1999, Shrikant Jagtap [ Ann - This is probably him, unless there are two Shrikant Jagtaps at UFl ] wrote:


I'm enjoying the current debate about CO2 levels. I feel that we are using the GCM scenarios, and we MUST use exactly those CO2 levels for crop model runs, so all data is consistent. So if we are wrong, we are uniformly wrong and adjust our explanations accordingly whenever we agree on things. Now to use different data will be hard to explain.


Dr. Shrikant Jagtap
104 Rogers Hall, Ag & Biol. Engineering
University Of Florida
Gainesville, FL 32611
This one gets a dramatic response:

Final e-mail:
From: Dave Schimel [Ann - this is probably him ]
To: Shrikant Jagtap
Subject: RE: CO2
Date: Mon, 17 May 1999 09:21:35 -0600 (MDT)
Cc: franci , Benjamin Felzer , Mike Hulme ,,,,, Mike MacCracken

I want to make one thing really clear. We ARE NOT supposed to be working with the assumption that these scenarios are realistic. They are scenarios-internally consistent (or so we thought) what-if storylines. You are in fact out of line to assume that these are in some sense realistic-this is in direct contradiction to the guidance on scenarios provided by the synthesis team.

If you want to do 'realistic CO2 effects studies, you must do sensitivity analyses bracketing possible trajectories. We do not and cannot not and must not prejudge what realistic CO2 trajectories are, as they are ultimatley a political decision (except in the sense that reserves and resources provide an upper bound).

'Advice' will be based on a mix of different approaches that must reflect the fact that we do not have high coinfidence in GHG projections nor full confidence in climate ystem model projections of consequences.

That one seems to speak for itself. Read it a couple times, because it looks quite important: don't use realistic scenarios because that's a political decision. If you do use realistic scenarios you "are out of line." That last paragraph is a whopper too: "we do not have high confidence in GHG (again, I think that's green house gas) projections, nor full confidence in the system model projections.

In other words: we're talking out of our ass, so don't start doing realistic data analysis and screwing everything up! Keep it fake, keep it inline with what we need.

Again, this doesn't exactly look like science, does it?

TKO for AGW? (II)

Here is a very important post on Watts Up With That. Much of the buzz over the hacked/whistleblower-released data from the CRU climate agency in Britain has focused on the e-mails, but there was actual data and computer code released as well. Now, people are beginning to focus on that. Here is some of the comments from one computer code:

; UNFORTUNATELY THE IDL5.4 p_correlate
pro maps12,yrstart,doinfill=doinfill
; Plots 24 yearly maps of calibrated (PCR-infilled
; or not) MXD reconstructions
; of growing season temperatures. Uses
; “corrected” MXD – but shouldn’t usually
; plot past 1960 because these will be
; artificially adjusted to look closer to
; the real temperatures
(Ann-Extra paragraphing added for readability.)
The first paragraph shows the level of tweaking that happens to get the program and the data to fit. If you have to do this, it's a good sign that your equations are crap.

The second paragraph holds the first bombshell, though: "these will be artificially adjusted to look closer to the real temperatures". In other words, the coder will massage the results later to make the model look more accurate for the more-recent time period. In other, other words: The objections people have had for years that these models can not accurately predict the past, were far more true than they even knew.

Here's another stunner from another code, which might be talking about the same 1960+ problem; a distinct echo of the "hide the decline" e-mail:
From documents\harris-tree\

; Computes regressions on full, high and
; low pass Esper et al. (2002) series,
; anomalies against full NH temperatures
; and other series.
; Specify period over which to compute
; the regressions (stop in 1960 to avoid
; the decline
(Ann-Extra paragraphing added for readability)
The ultimate question has to be: is this really science? Because it sure looks like they were making it up as they went along.

Default is a four letter word

Here's a scary statistic:
[ NYTimes ] With the national debt now topping $12 trillion, the White House estimates that the government’s tab for servicing the debt will exceed $700 billion a year in 2019, up from $202 billion this year, even if annual budget deficits shrink drastically.
Keeping in mind that these things always turn out worse than expected, and that it assumes "budget deficits shrink drastically", this is shocking. 700 Bil was the size of the entire federal government budget the year I was born. In 40 years, we've gone from that, to that merely being our mortgage payment.

Let the fledglings fly!

I've heard of over-parenting in college, but this is stunning, pathetic, ridiculous, and so many more adjectives I can't think of right now (just had some tequilla).
[ Time ] Overparenting had been around long before Douglas MacArthur's mom Pinky moved with him to West Point in 1899 and took an apartment near the campus, [...] But in the 1990s something dramatic happened, and the needle went way past the red line. From peace and prosperity, there arose fear and anxiety; crime went down, yet parents stopped letting kids out of their sight; [...] By the time the frenzy had reached its peak, colleges were installing "Hi, Mom!" webcams in common areas, and employers like Ernst & Young were creating "parent packs" for recruits to give Mom and Dad, since they were involved in negotiating salary and benefits.
These should be fully-fledged adults, pioneering their way through the first-real-job jungle, machete in hand and and just enough rations in their ruck sack to get them to Wednesday. Instead their parents are acting like their kids are 3-year-olds on the first day of preschool!

And, I must say, what's wrong with the kids-cum-adults?! Isn't this what the teenage years are all about? Isn't that when you distance yourself from your parents, assert your independence and demand to be treated like an adult ( who still has mom do their laundry and the rents pay for tuition, of course.) How can any self-respecting 21-year-old stand to have this kind of oversight? It's one thing to consult with parents when dealing with such a brand new world as employment contracts, but the extent of the dependency this article suggests is deplorable.

Read the whole thing. I'm generally in favor of the Free Range movement, I have Leonore Skenazy's book and I'm a regular follower of her blog (See just to the right here for the link permanently installed on our links list -->). Our only problem around here is the plethora of bums. When you have a strung-out guy living at the bus stop on your corner, you're a little wary about letting the kids out the front door (especially the 9-year-old girl).

(As a completely non-sequitur aside: I knew the Time magazine had a cover story on over-parenting, so I went to their website to try to find it. It took me a while to do so. You would think that the story that they decided would sell the most magazines this week would be prominently displayed on their website. Instead, it was third on a list on the bottom half of the page, and took me half a minute and a double-check that it wasn't last week's cover story to find it.)

(As a second completely non-sequitur aside, this is a quote from a little later in the same Time article:

"Eleven parents are sitting in a circle in an airy, glass-walled living room in south Austin, Texas, eating organic, gluten-free, nondairy coconut ice cream."

You know, I'm pretty darn sure that all ice cream is gluten-free. At least, I've never seen a recipe that calls for flour!

So, my question is, is it the parents who are dorks for insisting on gluten-free ice cream, or the reporter for writing such stupidity?)

Parody or not parody?

I can't quite tell if this is a real video by an environmental group, or a parody of what sort of videos environmental groups put out:

The website listed on the video: Plane Stupid could also either be real or parody. The subhead on the website is: "Bringing the aviation industry back down to earth." This is their logo:

I'm leaning toward must be a parody, right?

Global Warming Calculator - II

A few days ago, there was an interesting post at "Watt's Up With That?" called "The Steel Greenhouse." The subject of the post was a two-layer atmospheric variation of the standard Kiehl & Trenberth energy transfer accounting that describes the greenhouse effect. I had previously created a "Global Warming Calculator" for the original K&T model, so I decided to do a similar thing for this new two-layer model. You can see the results here: "Global Warming Calculator - II." The only tricky part was coming up with closed form formulas for the steady-state levels of radiation from the various levels. If you use the "view source" feature of your browser, you can see that the formulas within the java script are pretty long.

It was just for fun. See here and here for previous related posts.

It's on!

The Large Hadron Collider has been turned on.

Which necessitates the posting of this link: Has the Large Hadron Collider destroyed the world yet?

TKO for AGW?

I've spent quite a bit of time today going around the web looking into the hacked data coming out of a major British climate research agency. Someone broke into their computer system and downloaded more than 61MB of zipped emails, data and computer programs. The University of East Anglia’s Climate Research Unit (CRU) is one of the premier pro-warming climate research agencies in the world, with much of their research showing up in places like the IPCC reports. It has also long been the target of skeptics who have been using the British equivalent of Freedom of Information requests to try to get a look at the underlying data they use in their research. That's a fundamental tenet of science--that data should be shared so that others can verify your research--the CRU isn't a big fan of that idea.

So far, most people have focused on things like the head of CRU seeming to describe in an e-mail how he massages data to get the right result. In one memorable e-mail he talks how to overlay different data sets to "hide the decline" in temperatures. Another e-mail, I believe by the same guy, talks about how inconvenient a blip of warming was in the 1940's and how they have to massage it downwards a bit in order to hide it better.

Other bloggers are focusing on the efforts the e-mails show to hide data and to do an end-run around freedom of information requests--including sitting down with the government officials in order to show them why it is important to understand that the requests are coming from skeptics and should therefore be blocked. If these allegations can be proved, then there is actually a possibility of criminal prosecution, because interfering with an FOIA is actually a criminal offense.

Still other bloggers are looking into what appears to be pressure being exerted on peer-reviewed journals to make sure that skeptics don't get published and don't get a voice in the IPCC reports.

Lot of people are following this.

Gateway Pundit (also here) has a good overview post with lots of links.

Climate Depot is tracking links. (Also here).

As always, Watts Up With That is tracking it with a number of blog posts.

This post over on The Air Vent has a list of MSM reports on the hack. Including Fox, UK's Guardian, NPR, UK's Telegraph, The Wall Street Journal, BBC, Nature, the New Scientist and NRO.

People are going to be pouring over the information from this hack for a long time, and the next few days should continue to be entertaining, and hopefully enlightening.

Update: A great list of links is available here.

Indian Summer

It was a beautiful Autumn day here in Nashville. We took advantage of it with a trip to the park. Living in the South is starting to pay off, weather-wise.

Health Care Bill Scare

Interesting post from one of the guys at JammieWearingFool: "Congress Voted To Fire Me Last Night"

Their vote in essence caused me, and just about everybody else who works for a health insurance company, onto the unemployment rolls. It doesn't matter whether you are a claims adjuster, work a call center, sell the policies or some other job and you work for a private health insurance company or you are one of the evil heads of the corporation, your shelf life for having a job if the Pelosicare bill becomes fact is about five years.

If you currently have a private insurance policy, starting in 2011 your premiums will double if you are lucky, and in most cases triple. It is unavoidable.

My advice to anyone who works for a private insurance company is to dig out the resume, update it and try like hell to land one of the government jobs.

Read it all if you dare! Another, related view here: "PelosiCare's price controls will bankrupt health insurers in one year"
The blog points to text that in the bill that sets the minimum allowable loss ratio at 85%

An 85% loss ratio, as mandated by PelosiCare, would bankrupt insurers within a year. No mandated loss ratio has ever come close to 85%.

Graph taken from Here.

Yipe! Is he right? The issue seems complicated. Here is an interesting paper (pdf)

ABSTRACT: This paper examines the use and abuse of the medical loss ratio in the contemporary health care system and health policy debate. It begins with a survey of the ways in which the medical loss ratio has been interpreted to be something it is not, such as a measure of quality or efficiency. It then analyzes key organizational features of the emerging health care system that complicate measures of financial performance, including integration between payers and providers, diversification of payers across multiple products and distribution channels, and geographic expansion across metropolitan and state lines. These issues are illustrated using medical loss ratios from a range of nonprofit and for-profit health plans. The paper then sketches a strategy for improving the public's understanding of health plan performance as an alternative to continued reliance on the flawed medical loss ratio. This strategy incorporates data on structure and process, service quality, and financial performance.

Green idiocy

I was watching a home improvement show which just happened to be on the Planet Green cable channel. On came a commercial where Steve Thomas, former host of This Old House, said a truly idiotic thing.

The commercial was quick, but while he appeared to be looking at either a furnace or a water heater he said of the no-doubt magnificent heater: "And this really does have a measurable impact on global warming."

Measurable impact?! Those were his words. "Measurable impact." I burst out laughing.

Utter and complete scientific illiteracy being presented as profound scientific truth.

A perfect portrait

This seems to me to be a perfect portrait of ousted Honduran president Zelaya:

That's him in a room in the Brazilian embassy, complete with two aero beds and tin-foil covered windows!!


I hated the original Stargate movie, but when I gave the SG-1 series a try, I liked it. Around home it's referred to as "Auntie's bad movie." I haven't been a fan of Atlantis, and watching a tiny bit of Universe, I don't think I'll be a fan of this one either.

You see, it seems to have a holodeck problem. The main plot of the show is that a bunch of people, military, science, and civilian, are stuck on a space ship umpteen light years from Earth. Plenty of possibilities for danger and dramatic tension as they struggle to survive. However, in the earlier SG-1 series they established an alien technology that allowed people from one place to inhabit the bodies of people far far away, to walk and talk from inside other people, all instantaneously--over thousands of light years. So right now, the people who are supposed to be stranded on a ship which is running out of fuel, air, water and food, are partying in a night club back on Earth. Blah!

Square Dots

I have a new post up on Square Dots.

First Friday Employment Post

Here's the latest employment graph:

It looks like within a couple of months, the number of people employed will be less than at the depths of the post-911 recession--this despite an increase in population of about 11 million since 2003.

It also does not look like discouraged workers are beginning to look for work again. The numbers of long-term unemployed remained about the same, and the number out of the workforce has increased since last month.

If there is any bright side, it looks like the slope of the establishment survey line is beginning to shallow out slightly, though the household survey has not done the same. And both lines are still on a steep downward trajectory. A rough calculation of the average of the establishment line from October 2008 to July 2009 shows a monthly decline of about 0.593 million jobs lost per month. The same calculation over the last three months, July 2009 to October 2009, yields an average of about 0.136 million jobs lost per month--still a decline, but the slope is leveling off. The household data shows almost no change, if not a slight uptick in unemployment. The earlier period saw an average loss of 0.51 jobs per month, the most recent 3 months show a 0.59 jobs lost per month.

Do you Hulu?

I admit to being a fan of the website Hulu, where you can watch TV with limited interruptions (sometimes none) and watch at your own time and pace.

It seems to me that TV programming execs should be scared out of their wits by this and the Tivo phenomenon. The simple reason is this: if you don't have to watch something at a specific time, you tend to find other, better things to do, and, unless a show is really good (a rarity,) you begin to watch it less and less often.

I stopped watching Heroes because of that phenomenon (that and the fact that they didn't learn the correct lesson from Joss Whedon's Buffy: since the villain is almost always the most-interesting character on the show, you should change the villian every year to keep the show interesting.) And, now I feel myself slipping away from "Glee" (though the Kristin Chenowyth episode was great! as was the dancing football team!)

NY-23 Results

I was right about what would happen in NY-23. The Republican who lost support and dropped out still got 6 percent of the vote, and that was enough to give the win to the Democrat over the Conservative Party candidate.

Today of course, you will have endless talk from pundits about the "deeper meaning" of this race, most of it nonsense.

Truth to Power

Student criticizes Khamenei at official gathering:
A lone student at a gathering of the country's academic elite took the unprecedented step of criticizing Leader Ali Khamenei in his presence on Wednesday morning, according to the Office for the Preservation and Publication of the Works of Grand Ayatollah Khamenei and opposition news sites.

Read the whole article. I am astounded by this student's bravery. I wish him well.

Economic recovery

I'd be very surprised if we were actually seeing the beginning of real growth in the economy, unless, of course, we are talking about a brief uptick followed by a second retreat. The reason is very simple:

Why would any employer of any size hire one additional worker in this climate?

Similarly, why would any prospective entrepreneur go out on a limb to start a new business in this climate?

Why would anyone take the risk of such things with so much uncertainty about the economic future and government mandates, regulations, and stupidity?

American Wins the 2009 NYC Marathon

Meb Keflezighi became the first American man to win the New York City Marathon since 1982 on Sunday.
"The USA gave me all the opportunities there is in education, sports and lifestyle," he said. "To be able to represent the USA is a big thing for me."


Score One for Hillary!

Hillary Clinton, during her recent trip to Pakistan:

"Al-Qaida has had safe haven in Pakistan since 2002," she finally asserted when challenged about Washington’s tough prescriptions for Islamabad. "I find it hard to believe that nobody in your government knows where they are and couldn't get them if they really wanted to."

The US Secretary of State also took a swipe at the Pakistani military and intelligence agencies, telling the senior journalists, "If we are going to have a mature partnership where we work together" then "there are issues that not just the United States but others have with your government and with your military security establishment." She said she was "more than willing to hear every complaint about the United States'' but the relationship had to be a "two-way street."

At one point during the exchanges, when a journalist spoke about all the services rendered by Pakistan for the US, Mrs Clinton snapped, "We have also given you billions."

Well done Mrs. Clinton! How refreshing to hear an American diplomat dishing it out rather than just standing there and taking it. Finally, someone in power has the nerve to tell the truth.

Bayes, Laplace and the Sun

William M Briggs mentions Laplace's Rule of Succession in a recent blog post. Briggs' is a blog about statistics and related matters that I highly recommend. Laplace used the rule, which relies on Bayes' Theorem, to calculate the probability that the sun will rise tomorrow. It is an elegant and fascinating bit of analysis. According to Wikipedia, Laplace's method give odds 1826250:1 in favour of the sun rising tomorrow.

But I beg to differ! Using Bayesian analysis, I calculate that the probability that the sun will rise tomorrow is 1/2!

Here is my argument. Let N+1 be the number of days from the start of history through tomorrow. Suppose at first that we know nothing about the sun, neither the related physics nor the past history of its rising. This was Laplace's assumption as well. Suppose we only know that the sun rises on some subset of the N+1 days in question. With only this knowledge, we assume a uniform prior probability distribution on this subset. Thus we assume that all 2^(N+1) possible subsets of the N+1 days are equally likely to be the sun-rising subset, each having probability 2^(-N-1). Now suppose we are given additional knowledge, specifically that the sun rose on the first N days. There are then only two possibilities for the sun-rising subset: the set of all N+1 days, and the set that contains only the first N days. By our prior assumption, and a trivial application of Bayes Rule, we see that each of these possibilities now has the posterior probability of 1/2. Thus the probability of the sun rising tomorrow is 1/2.

Put that in your pipe and smoke it! Related post here.