Deinonychus antirrhopus: terrible claw

Saturday, May 17, 2003

 
Good Bye Blogger Yep, I've had enough. Dammit, I am sick of blogger an all the ups and downs in terms of bloggered permalinks, bloggered archives, and just not being able to log into blogger to make a post. And, please lets not even start on the number of times blogger has eaten a lengthy post.

So you can read my bloviating at...

Deinonychus antirhopus, a.k.a. http://www.steveverdon.com

I'm still working on the site, but I think it is at a stage where I can start using it as my main blog and keep this site as a sort of backup.
Steve 11:08 PM

 
Social Security Reform: Who Are the Real Bad Guys One of my favorite examples of the left's/Democrat's own visciousness is their portrayal of those who want to reform or even get rid of Social Security as some sort of monsters who want to throw the elderly into the street to be run over by a garbage truck. (Link, link, link, ) Not all responses use that kind of rhetoric (link) although there was that horrendous cartoon on the DNC's website of Bush pusing elderly people in wheel chairs over a cliff. But who really are the bad actors here...?

Lets whip out some of the tools of economics and look at this a bit more closely. Now in economics most people are exposed to the static models, i.e., models that are a "snap-shot" in time. These models have no dynamic element, and hence there is no need for savings. Now, in dynamic models not only do you make decisions in the "current period" such as how much of a given commodity to purchase, but you also make decisions in the current period that affects later periods, name savings. Now, lets also not forget that in economics people do things that they percieve are in their best interest (maximize utility subject ot various constraints). So what ever you are saving is what you think is best for you given your situation.

So lets suppose we get rid of Social Security, and further we manage to come up with a way so that those already recieving benefits or who are about to recieve them are not hurt (i.e., they are paid off, after all these people made plans assuming Social Security would be there so it is unfair to take it away so "late in the game"). Now, are there any arguments you can make against this? Not very good ones, IMO.

1. People wont save for their retirement

My resposne is, "Yeah so?". Remember these people are making decisions that they percieve to be in their best interest. That is, if they are not saving for later it is probably because they need to spend the money right now. Think about that for a minute. Who is least likely to save for a rainy day? Those at the lower end of the income distribution? These are precisely the people who would find a few extra bucks at the end of the month very helpful in terms of covering expenses. Sure it sucks that it is a hand-to-mouth existence, but are you so sure of the moral superiority of your views that you are going to take those few bucks and give them to somebody else?

Yep, you read that last sentence right. Social Security is not a retirement plan, it is a transfer program where it takes money from those paying in today and pays out to those who are now eligible. So Social Security really isn't even providing for their future, it basically provides for somebody else's "right now". The plan does intend to be around so that when those working today reach eligibility they can get money from those people who are working and paying into the program at that future date, but it is most definitely not a retirement plan.

2. People don't always make the best decisions

Two resposnes to this one, "How do you know?" and "Again, so what?" The first answer I base on the fact that when people often say this it is based on hindsight and their own personal preferences. Hindsight is always perfect, "Oh if only I had known that was going to happen I'd have done things differently..." Yeah, but you don't always know what is going to happen, do you. In economics, this is captured by introducing uncertainty into the models, random variables and having the consumer/individual using such concepts as expectations to make decisions. These expectations are conditioned on information so those with more information are probably going to make better decisions. Still, this doesn't mean that people are making dumb or bad decisions, they just can't predict the future perfectly.

The second response follows from the idea that these people are adults and are free to make their own decisions. That is, lets suppose that it is true that people make dumb decisions, shouldn't that be allowed? There are two things here to consider, IMO. The first is the notion of freedom. That is people should be free to make their own decisions, even if they are (from your perspective dumb). Unless that decisions impacts on another person, there really isn't a reason to stop them from making that decision. The second factor is one that economists call "perverse incentives". By always shielding a person from the fall out of their "dumb" decision making you basically take away the incentive to not make dumb decisions. If we want people to save for their future retirement/old age, forcing them into a program like Social Security will blunt the incentive to save the future. Note that this blunting of incentives is even more true if we assume that people are rational (i.e., making the best decisions given their information).

3. Its provides a safety net

Not in its current form. As soon as you hit a certain age you are eligible no matter your wealth. Further, even if we changed it so that it really was a safety net, then again you have the problem with perverse incentives. Those who are border line in terms of whether or not they will be recieving Social Security will now have less incentive to save, thus it will be an overly large safety net. Also, as the net is increased to provide a safety net for those with higher incomes, those who are "borderline" also moves up the income distribution. It is possible that the safety net could grow quite large.

So, it seems to me, that at the very least here those who oppose finding a workable solution to Social Security in its current form really are on the wrong end in terms of the moral aspects. It takes money from those who could use it the most, it presupposes people are not smart enough to make good decisions (given current information), and it is another example of the condescending Nanny State mentality that people are really big children who need to be taken care of. Yet, time and time again it is the Democrats howling about the immorality of any suggestion that Social Security be changed. Perhaps the volume and level of vitriol is because deep down they realize the bankruptcy of their own position on the issue?
Steve 9:58 AM

Thursday, May 15, 2003

 
Dean's Demagouging Howard Dean as promised that health care will be the center of his 2004 presidential campaign. Great, just what we need, HillaryCare Part II. Anyhow, not content to simply put this policy notion on the table for discussin Dean has to run his mouth off and engage in some might fine stretching of the facts

"Here, in the richest, most advanced country in the world in the 21st century, it's simply wrong for a sick child to go without seeing a doctor because her parents can't afford it,'' he said, noting the United States was the last of the major industrialized countries to provide universal health care.

Yes, it would indeed be ashame if it were even true. Sadly, for a putz like Howard Dean this is not the case. I am so sick of this silly canard always rearing its head time and time again.

Mother: "Little Johnny is gonna die without the heart transplant and we just don't have the money."
Father: "Well lets try to make his final days as wonderful as possible..." *sniff*sniff*sob*...

Yep, every hosptial loves to be the hospital that wheels little Johnny out into the streets on a snowy, rainy, sleeting night on a gurney without even a sheet to keep his poor tortured body warm.

Apparently Dean uses that same bizzaro world definition of "child" that many Democrats love to use.

His proposal would expand existing state and federal plans to allow health care coverage to every child and young adult up to age 25 while offering tax and other incentives to help working families afford insurance.

See, the Democrats (or at the very least those who like Dean) see even adults as nothing more than great big children who need to be taken care of by the wise goobermint. And what the Hell is a "working family"? Does this mean that if you have an income over a certain amount you are not really a "working family"? Ahhh, nothing like a little class warfare rhetoric to toss in with your socialist mumbo-jumbo.

Dean aides estimated the plan would cost about $88 billion annually, and Dean said he would pay for it by eliminating portions of Bush's tax cuts.

What a pip that Dean is. He should take his act on the road, "How to spot a liar." Let me translate that above crap for you, "You the tax payer will pay for it." That is basically what Dean is saying, of course being a politician he can't come out and say that. Instead he talks about how he will pay for it.

Gephardt also decided to get in on this act as well

Democratic Rep. Richard Gephardt of Missouri outlined a sweeping universal coverage plan last month that he said would also help stimulate the economy by easing the health insurance burden for employers. His plan would cost more than $200 billion annually, which Gephardt would pay for by repealing Bush's tax cuts.

Yeah, and in your garage Dick is a little pony for you to ride around on. For the love of God, what is it with these nitwits. Lets assume that Dick's plan has no administrative costs (i.e. bureaucrats aren't sucking up a big chunk of that $200 billion--would you stop laughing Robin). Now, that means businesses spend $200 billion less on providing healthcare to their employees. But, the money to pay for it comes out of taxpayer pockets. So, you really aren't going to get more economic stimulus economy wide, you will just be moving money from one actor in the economy to another.

And lets not forget the deficit. Remember how all these Democrats are railing, nashing their teetch, and having hysterics about the deficit, yet they seem more than willing to run a deficit through increased spending. If the deficit is bad because it pushes up interest rates or something else, it is bad. Major inconsistency problem here.
Steve 1:20 PM

 
Interesting Post on Medical Care at Asymmetrical Informaiton The post debunks the notion that people without insurance are not aggressively treated (in a medical sense). The comments are also worth a look.
Steve 12:40 PM

 
Jacob Levy on Taxes Not a bad post overall, I agree that the current tax cuts proposed by President Bush are iffy on the issue of short term stimulus, but one thing that jumped out that I am going to pick on.

I think it would be better to have a coherent supply-side tax cut (say, a small permanent reduction in the capital gains tax) or a coherent Keynesian cut (a three-year increase in personal exemptions) or a combination of the two or... well, anything that at least might be intelligible.

The italicized part is not a good idea for stimulus, or at least not the best idea. You can get the same dollar amount in with cuting tax rates, the benefit of this is that you'll also be improving the after tax wage so that people will want to supply more labor. More labor supplied means that you have more things produced, which means you'll not only have people spending the extra money from the tax cut (which is what you'd get from the increase in personal deductions), but you'd also get people spending more money from the additional labor income they earned (which you don't get with the increase in personal deductions).

Lots of people like the idea of increasing personal deductions, but it is actually less effective than cutting tax rates on income (which for many Americans is labor income), IMO. Why this is so I haven't go a clue. It isn't like increasing the personal deductions by $2,000 means you are going to get an additional $2,000 in after tax income. Oh well, who knows...maybe people just get confused my the notion of marginal effects.
Steve 12:20 PM

 
Voting Booths vs. the Market Place An interesting article on the problems of relying on the political process to solve problems with the market. There is one interesting point the article raises:

Retrospective studies like Peltzman's gloss over a serious logical flaw at the heart of regulation's popular rationale. Once the flaw is recognized, the rationale crumbles like a house of cards. The point is simple yet powerful. To wit: in democratic societies, elected political officials are the final arbiters of the government's marketplace do's and don'ts. Who elects these political officials? Surprise! The same people supposedly incapable of making informed decisions in the marketplace.

Which is one reaon why we have bureaucracies. The politicians cannot hope to be well informed on the myriad of problems, issues, products, and services that are offered in various markets. So, they have the bureaucrats jump in and do it. The problem is that even with the bureaucrats the politicians are still woefully ignorant of many issues. Such as suppose you have a terminal disease, you basically know you are going to die sooner rather than later. Further, lets assume there is a drug out there that might put off death. Some poeple might rationally decide to try this drug, even if all of its effects are not yet known in the hopes of living. Typically most regulations prevent such decisions.

Lets try an example, suppose you are sitting there watching two people, one is drowning the other is standing there with a pole watching him drown. You say to the man with the pole, "Your pole is long enough to reach him, why don't you put it out for him to grab onto and we can pull him to safety." The drowning man is yelling basically the same thing. The guy with the pole sadly shakes his head and says, "Well, that does sound like a good idea, but if I take your and his advice, I might hit him accidentally with the pole. I could knock him out, thus killing him or just poke an eye out or something like that. Sorry, I think it is safer for that guy if I don't try anything."

Sound stupid? Welcome to the world of many regulations. We don't care if you realize you could die. We don't care if you want to take this drug knowing you face unknown risks. We know what is right for you, and we are going to make sure you do what we know is right for you.

Now this doesn't mean I want a return of the "good ol' days" when there were snake oil sales men either. I think there is a role for the government, specifically the role of providing information for very low cost or even for free (well not free exactly, but funded by taxpayers, but no charge on obtaining the information). Obtaining information can be costly, the government can help reduce this cost (or more accurately spread the costs around more) and thus allow more people to make more informed decisions. This isn't a perfect solution, but then agian neither is the solution we have now.
Steve 12:06 PM

 
Misquote of the Day Eugene Volokh catches Maureen Dowd in...well a lie. She mangles a quite by President Bush to fit the message she wants to send to her readers and basically distorts the facts. No wonder I don't read her drivel.

Correction: Henry Hanks notes that it was Andrew Sullivan who caught Maureen Dowd.
Steve 10:59 AM

Wednesday, May 14, 2003

 
Simply Staggering For years this dimwit worked to get the sanctions agains Iraq lifted. Now, when it looks like the sanctions are going to be lifted which will help the Iraqi people...sliming onto the stage is this guy who

After five years spent working to end the sanctions on Iraq, I find myself in an odd position. I'm opposed to the current U.S. plans to end the sanctions.

Odd? Let me see....he didn't like the sanctions imposed by the U.N. because Iraq had a nasty tendency to develop weapons of mass destruction, and even use them on occassion, but now...NOW the sanctions are a good thing. Why?

Today, the United States is willing to (partially) withdraw after it installs its own puppet regime (one that will presumably have more independence than the one Iraq tried to install, but will still be subservient to U.S. dictates). It also wants to plunder Iraq's oil wealth for its own political purposes and for the benefit of U.S. corporations. This is reason enough to keep the sanctions on until there is a legitimate Iraqi government. This can only happen if U.S. and other "coalition" forces withdraw, there is a multinational U.N. peacekeeping force with no participation from any of the aggressor nations, and the Iraqis are given a genuine chance to exercise their right to self-determination.

Now, lets grant the dimwit here everything he claims, i.e., the sanctions were harsh on the Iraqi people. Now he wants to continue this harsh treatment because he does not want a puppet regime in Iraq that will do the bidding of the U.S. Is it just me, or has he just become exactly what he supposedly reviles? What a freaking dimwit.

Once the regime is destroyed, miraculously, the Bush administration realizes overnight that sanctions were actually harmful and that it's necessary to remove that burden from the Iraqi people in order to provide humanitarian aid and reconstruction.

What a wonderful bit of misrepresentation here. Of course both Bush Administrations, the Clinton Administration know/new that the sanctions would impose a hardship. That is the point, we will do this which will impose a hardship until you stop this "bad" behavior. Now, that the primary driver behind the bad behavior is gone it is time to lift the sanctions so that the Iraqi people will no longer have to suffer.

My guess is that this guy just hates everything the U.S. does. It wants to impose sanctions...that's bad. Now the U.S. wants to lift them...why that is bad too. The U.S. wants to destroy its nuclear arsenal...that'd be bad too, because without the U.S. to despise I think this guy would not know what to do. But that's just a hunch.
Steve 2:50 PM

 
The Evil Oreo Cookie A lawyer in California wants to ban the sale of Oreo cookies because of the trans fats that are used in the making of Oreos. What always gets me about these nanny types is that they always point to what medical research says. You are x feet and y inches tall therefore you should weigh z pounds. The problem with this (besides the problems with such generalizations on scientific grounds…i.e. what if you are big boned, are you supposed to adhere to that weight or weight range?) is that it ignores what people want. People might want to be fatter than what the medical research says.

See, this is the way I think of it, there is the quantity of life issue. That is, by maintaining a good diet and “proper” weight you’ll probably live longer. There is also a quality of life issue as well in that your later years will not be filled with as many medical problems (on average). However, there is also the issue of enjoying life right now, and part of that is eating good food. And by good food I mean food that tastes good, and generally speaking food that tastes good is food that has fat and is probably bad for you (I loathe brown rice, it sucks it tastes like warm wet cardboard). So you have these various trade-offs that you have to make decisions about. Do you want to enjoy the here and now more than the (much more) uncertain future? Do you want to be able to do more physically?

Not everybody is going to make the same decisions here, the decisions depend on a host of factors that vary across individuals. Yet what we see from these nanny types is a desire to come up with a “One size fits all policy”. It’s bad so you can’t have it anymore. These people are just like the Prohibitionists back in the 1920’s and 1930’s that got alcohol outlawed. That sure worked well didn’t it. Funny how often these same nanny types will blabber on and on about diversity and how it is good for you, but when you have a diversity of preferences and thus a diversity of outcomes (in this case a diversity of weight) it is suddenly bad. Everybody has to adhere to what the body mass index says their weight should be. (If it hasn’t dawned on you yet, inconsistency sort of annoys me.)

Further, to make this even more annoying is the good intentions of these kinds of people. “It’s bad for the children, and we have to protect them.” This is truly annoying on several levels. First, it assumes that all parents are bad parents in that they are not regulating what their child eats or drinks. Granted some parents probably do let their children eat too much candy and cookies. Still, is it necessary to punish those parents who are responsible and let their child consume Oreos in moderation? I don’t think so. Another reason it is annoying is the authoritarian nature of these busy-body do-gooders. They have decided something is too unsafe for you or your children and so they will now make a decision that denies you access to this something. This isn’t plutonium or heroin we are talking about, but something that might kill you if you manage to eat too much of the item. Finally, these types of people just assume the rest of us are just too stupid to be making our own decisions:

The big difference between this suit and others that have targeted tobacco and McDonald's fast food is that consumers know that tobacco is bad for their health and that McDonald's food contains a lot of fat, Joseph said.

"Trans fat is not the same thing at all. Very few people know about it," he said, explaining that his suit focuses on the fact that trans fats are hidden dangers being marketed to children.

Excuse me? Sorry there Mr. Joseph I realized almost 20 years ago that Oreos were probably not the best food on earth from a health/weight perspective. I still liked them, but I knew that they were quintessential junk food and not something I should eat in large quantities. The idea that people are ignorant of the detrimental impacts of Oreos or other products is quite telling of the massive arrogance and ego that these types of people exhibit.

These types of people, if given enough leeway will want to control a great deal of people's lives. Don't eat this, eat that, exercise this way, use this, don't use that. These people are control freaks, and they should be vigorously opposed, IMO. At the same time, I don't have a problem with informing consumers as to the deleterious effects of various ingredients such as trans fats. That is fine, in fact I think that is wonderful. Giving people more information allows them to make more informed (and thus generally better) decisions. But notice the difference, I am in favor of still letting people make decisions, proto-fascists like Stephen Josephs is not, he wants to be the one that makes decisions for a large number of other people. Mr. Joseph...get the Hell out of my kitchen.
Steve 11:35 AM

Tuesday, May 13, 2003

 
Arianna Huffington I just read that and I gotta say, she is one whacky woman. Not all that long ago she was rather conservative, now I don't know how to describe her other than just whacked in general. Still the blind chicken does occasionally find the kernel and in this case she has a point. If anything is going to torpedo Bush's re-election bid it is going to be the economy. He needs to get it turned around and turned around now or if later in a way that is clear to most people. That was the problem his father faced. The economy was actually expanding months before the 1992 election, but Clinton and the Democrats made it seem like the economy was still having problems. The recovery was anemic and many people just didn't know the economy was actually expanding. Bush's tax cuts might work, and personally I don't believe all this doom and gloom bravo sierra the Dems and other Lefties are blabbering about. In terms of % of GDP the deficits are not anywhere near as big as these chicken littles would have us believe. The only question is will it be enough and will it get the economy going soon enough.
Steve 3:55 PM

 
Busting Five Environmental Myths Thanks to Lynne Kiesling for the pointer. Anyhow, Ed Ring busts 5 common myths of environmentalism today. The myths are:

1. Being environmentally correct requires a lower standard of living.


2. Any good environmentalist is a socialist.


3. Hydrogen and other renewable energy are the answer to our energy needs.


4. We have to recycle everything.


5. New housing developments must be limited to within existing cities.

Check it out.
Steve 11:37 AM

 
Kevin Drum's Strawman Kevin points out that nepotism is more rampant than affirmative action (I suppose, although he doesn't provide anything to really back this up), and that conservatives who oppose affirmative action ignore it. What the...?!?! Why does he say this? Does he have any examples or quotes or anything? No. I mean a conservative saying, "Affirmative action is bad, but nepotism is different...," or something like that he might have something.

When you read a story about a decline in circulation at the New York Times, for example, does Mickey Kaus immediately speculate that it's because Arthur Sulzberger was appointed publisher instead of other more qualified non-Sulzbergers? If he reports that Ford Motor Company is perilously close to bankruptcy, is his first thought that this is a result of promoting Bill Ford to CEO instead of a better qualified non-Ford? If conservatives truly believe in a meritocracy, why aren't they busy denouncing this kind of thing, using their bully pulpit to shame rich whites into stopping this practice? Why indeed?

Nice isn't it. A well constructed strawman is a beautiful thing since it allows the person constructing it to so easily knock it down. Here's the deal Kevin, I don't give a rat's butt what a private company does (unless I am a shareholder) and neither should you. It really isn't any of your business, now is it (unless you are a shareholder)? I mean this kind of argument is facile and nonsensical. How about this, Kevin is married, when he goes out to buy a gift for his wife does he stop to consider another person in his life might be more desrving of such a largesse? No? Nepotism...rank and evil nepotism. Gee, guess I'll have to open up the process for deciding who exactly gets my son's college fund, after all he might not be the most qualified person.

Affirmative action on the other hand is not something that private firms and individuals institute it is something that is forced on them by the state. You will comply with this or you will be punished. Nepotism is not mandated by law, and is generally frowned upon. Generally speaking, no I don't like nepotism and think overall it can be a bad policy. So not only is Kevin wrong, IMO, that conservatives ignore this (although I don't consider myself a conservative), but also that the two issues are not the same. They are quite different. Just as it is differrent if Joe Public decides he doesn't like blacks, hispanic, whites or whatever vs. the government instituting laws that place one group at a disadvantage. That is Joe's right. I personally disagree with it and would not associate with Joe if I knew this aspect of his character, but it is his right to hold such disgusting views.

I'd dearly love to see what sort of programs Kevin would put in place of affirmative action, but sadly he decided for the rest of us that we are not interested, so I guess we can add arrogance on top of poor logic.

Update I: You know this reminds me once of an argument I had with a liberal co-worker back in my graduate school days. I pointed out that Clinton's desire for a female for a given post (Attorney General, IIRC) was basically sexism. I used the following analogy. You have a function f and lets suppose f is continuous on the set X and that we also have set A. Now it is clearly the case that

max_{x Î X} f(x) ≥ max_{x Î A} f(x).

Clearly since A Í X the inequality is valid. Now think of A as the set of all qualified female candidates and X as the set of all qualified candidates whether male or female.

Now, if a private company wanted to use such a decision rule (i.e., looking for females only) fine with me, but when it is the government doing it (hey if you don't necessarily want the most qualified person fine with me...well unless I'm a shareholder, then I might care), it is another thing, IMO.
Steve 9:58 AM

 
Whoa! Kloognome just got a make over. Go take a look, looks much cleaner, IMO. Dang, I really need to find/make the time to get off blogspot.
Steve 9:15 AM

 
Capitalism I often hear many on the left refer to captialism derisively, and it always makes me wonder why. Before we go any further though, I think it is important to define what I mean by capitalism. By capitalism I mean economies characterized by having the market as the primary method of exchange. There that was simple. Now one reason I find it curious that those on the left often refer to capitalism in less than glowing terms is because of the fundamental nature of markets. Markets are characterized by what economists call voluntary transactions. Now perhaps one possibility is that the term voluntary transactions is too opaque to these people on the left (note the wording, I fully realize not all those on the left have a disdain for markets). Well let me provide a translation (i.e., I'll explain it too you). It means that people get together and buy and sell things to each other because they want too. Nobody is forcing them to do it.

Now this begs the question, is there any other economic system that has such a characteristics? Command and control economies clearly don't becuase the government controls the resources and the means of production and commands how those resources and means of production are used, hence the name command and control economy. Now lets think about this for a second. To control and command the resources and means of production also means that you control and command labor, as labor is one of the factors of production. This means that the State controls the person at a fundamental level. Now in case I have lost you, this is just a fancy way of saying command and contrl economies are basically slave economies.

Now think about this for a minute. What was the South (here in the U.S.) like prior to the Civil War when slavery was still practiced? Was the economy characterized by industry and production? No, it was almost predominantly agricultural. The reason for this is quite simple, IMO. When you are a slave there is no incentive to do anyting other than the bare minimum that is required. If you work hard are you going to get a promotion? No, not really. If you work hard are you going to become free? No, probably not. Slave economies are typically low skill economies. Now think about places like Cuba, the former Soviet Union, and even North Korea and China. Are they particularly high tech? No, not really. So there is definitely a correlation here, and I think there is also cuasation.

So that lets out command and control economies, are there any others? How about fuedalism? Yeah, lets return to the days when the bulk of the population were serfs. Hmmm, not alot of freedom there. I suppose you could have some sort of in-between system where the government controls some of the factors of production and the private sector the rest, but is partial slavery any more desirable? Seems to me the partial answer is still going to blunt incentives.

So yes, I like markets, capitalism if you will. I realize that it is not the best system out there, but it is the best we have at the moment. I also find it curious that people who espouse systems that are basically command and control also have to talk endlessly about how much more free people are under those systems. It is almost like they are trying to compensate for something.
Steve 9:08 AM

 
Freedom Fries, the French and Germans I haven't really blogged about this, mainly becuase I think it is just sort of annoying and silly. However, I must confess that the most annoying people are those who run around scolding Americans who are peeved at the French and Germans for their actions (or more accurately inaction). After all, the Americans running around who are peeved at the French and Americans and who call french fries freedom fries, and talk about not buying German products are part of a backlash. Now, typically a backlash is a response to some prior offense, and in this case, IMO, the offense is not simply the French and Germans saying, "Hey sorry we disagree on this issue of Iraq and regretably we will not take part in such an action." No, instead it was rather heavy anti-American rhetoric and a resulting increase in hostility towards Americans in general (if the press reports are to be believed). So a hostile response, while perhaps not the most sophisticated response is somewhat understandable.

Does it bother me that Pratt-Whitney lost out on an airplane contract? Yeah a little bit, but lets not forget this is also going to hurt the French too. After all, now they are going to have to spend more money per airplane engine which means people flying on French airlines (i.e., mainly the French) will have to pay higher ticket prices. If they want to be vindictive to the point of stupidity, hey be my freaking guest.
Steve 8:50 AM

Monday, May 12, 2003

 
Gripe, gripe, gripe I read this and laughed. Sounds like Ted Rall is a bitter old man at such a young age. Some parts of it were hilarious, IMO:

Life isn't fair to the Democrats. No matter how much they suck up to corporate CEOs, they can't compete for contributions with Republicans who invite their backers to write legislation.

Well there you go you nitwit, just tell those CEOs you are busy sucking up to that you'll let them write their own legislation. Sheesh, problem solved.

Michael S. Dukakis served with honor in the U.S. Army for two years. Three decades later, he was ridiculed for riding in a tank while wearing a helmet and a goofy grin.

That is because he looked like an idiot, and also it was an obvious campaign ploy to try and convince people that Dukakis wasn't going to be soft on defense. Everybody knew he was, everybody knew the advertisement for precisely what it was, a blatant attempt to spin something. Bush's landing on the Carrier on the other hand was not an advertisement, he didn't look like an idiot, and everybody already knows Bush is not soft on defense.

Most registered voters are Democrats, but too many are disloyal swing voters and apathetic no-shows to assure victory.

Yes, those disloyal bastards...why they are positively un-American. Uhhh, wait isn't it a bad thing to say that...oh my bad Ted's a Democrat he can say whatever he likes.

And even when Dems do win the most votes, cheating Republicans bully their way into office.

Wow, somebody pissed in his cheerios this morning.

Post-attack America is feeling mean. We've used our pain as justification to kill thousands of Afghans and Iraqis, but we still haven't touched the bastards who hit us on 9/11.

Yep, he nailed it here. When I found out about 9/11 I immediately called my wife on the cell phone and told here what happened, that she shoud call work to see if she still has to go in, then I said, "Dammit, this is great! This means we'll get to kill thousands of Iraqi's and Afghans!" What a schmuck.

As for the bastards who hit us, yes we have touched them. Several of the leaders are dead, some are in custody, and some are on the run and if they stick their heads too far up out of whatever anus they have crawled in, it'll hopefully get blown off.

The recession has killed off millions of jobs, and no one seriously believes that Bush's tax cuts will bring them back any time soon. Liberals watch with dismay as the Bush Administration guts social programs, environmental laws and civil liberties; conservatives beat on lefties for expressing disunity during a time of crisis.

Yeah, like calling them disloyal like Ted Rall. As for social program being gutted I'd like to see which ones exactly are being gutted. Same thing for environmental laws and civil liberties. I'm not talking about little things, but gutting. Yes, yes, I am sure a few are going to point to the Patriot Act, but I'll leave it to Robin to point out the silliness of that one.

In theory Democrats should be able to beat Bush on the economy, but they won't. Like Bush, President Reagan ran up staggering deficits during his first term. He likewise refused to create jobs or to stimulate the economy during a deepening recession. But thanks to the pageantry of incumbency and a few sound bytes, he won a landslide reelection over the decent but dull Walter Mondale.

Would somebody please send this guy this link? The recession Rall is talking about ended in the fourth quarter of 1982 (November to be exact) which gave Reagan almost 2 years until his next election meaning that when people voted for him the economy was growing.

Go after Bush's ultimate Achilles' heel: run countless loops of the inarticulate Resident's clashes with the English language. "Too dumb to talk," a sinister voiceover reads. "Too stupid to trust." Use time-proven Republican methods, like name-calling: Extremist. Out of touch. Tax and spender. Hates workers. Racist. Homophobe. Corrupt CEO coddler. Idiot. Drunk. Cut to the post-pretzel-incident photo: "America needs a sober president."

Great idea, the Republicans can respond with such time honored Democratic name calling as: Racist, homophobe, cruel, evil, and wants to kill old people. Oh and somebody who knows Ted should get in touch with his family or something, Ted is in such a lather he got mixed up near the end there and resorted to time honored Democratic name calling techniques (I mean come on, if they are going to use Republican names, call Bush a homosexual maybe that way they can drive a wedge between Bush and the Christian voters!).

Geez, and they say those on the right are mean spirited. (Link via Jane Galt)
Steve 1:44 PM

 
Jane Galt on the Laffer Curve An interesting post about high taxes in New York City being pushed higher and its potential for long term damage to the New York City economy. The basic gist is that the Laffer Curve effect of higher taxes decreasing tax revenues as a result of businesses and people leaving the state is an excellent one. One that the bumbling incompetents in Scramento should keep in mind when they ponder what to do with the States horrendous budget problems. Frankly I doubt those idiots could keep much of anything in their anencephalic heads.

Oh and in case you are wondering if I think the politicians in Sacramento are idiots...you're right, I think they are idiots.
Steve 11:44 AM

 
Un-Fricking-Believable Up at the top of my blog was a link to Ironwood Publications. It had some blurb about how 400 economists include 10 Nobel Laureates had signed a letter warning against Bush's tax cut. Curious I clicked the link. I got to the page linked above. Here is what it said at the top:

The American economy, when operating properly, is one of the marvels of the world. It is a highly complex and abstract thing that almost defies description. Obviously, one cannot draw a picture, or take a photograph, of the economy. Yet, it is something very real.

As incredible as this gigantic mechanism is, it is not perfect. It requires constant attention to keep it operating properly. It sometimes moves too fast, and at other times it moves too slowly. Thus, policy makers must frequently tap on the "brakes" or push down on the "accelerator pedal." This marvelous mechanism has no automatic pilot control on it. Thus, even when it reaches the point where it is operating at the maximum efficiency level, there is no guarantee that it will continue to operate in the proper manner for very long.

Thus, it is necessary to constantly monitor the economy and take corrective actions when necessary. And it is absolutely crucial that those individuals who make decisions regarding adjustments to the economy have a sound understanding of how it works. Just as an inexperienced mechanic can do great damage to your automobile, economic policy makers who do not have a sound understanding of economics can do great damage to the economy. If the brakes are applied too hard, or the economy is accelerated too rapidly, it can take a long time for the economy to recover from the resulting damage.

I am simply amazed at the internal problems with consistency here. There is this great big marvelous economy that is so big, so complex, that nobody can really understand it...it is practically beyond comprehension. Practically, because this wonderful economy needs constant monitoring, constant fiddling with or else this wonderful marvelous, and practically incomprehensible thing will run amok. Doesn't that strike you as a little...well schizophrenic.

Its so big it is beyond comprehension...but if we moniter it constantly why we can control it. Uhhhh yeah. Sure, just like your care engine you merely have to know how and were to fiddle with these precise controls and soon the American economy will be purring like a high performance automobile. Never mind that consumers, firms, and other entities participating in the economy will do things you didn't expect which can completely prevent you from achieving whatever goal you have decided is good.

Plus the add for the book that promises to demistify the economy is a hoot. Its really big, and really abstract, and well practically beyond comprehension, but read this book and everything will be clear and you will no longer be bumbling around in the dark when it comes to understanding the U.S. economy. Somehow I doubt it.

A little further down there is this introduction to the screed against Bush's tax cut proposals

President George W. Bush's latest tax-cut proposal, like the massive cuts already enacted, does little to stimulate the economy in the short run, but it does a great deal to help the super rich at the expense of America’s future. According to the official statistics of the United State Treasury Department, the national debt rose from $5.728 trillion on January 22, 2001, two days after Bush became President, to $6.460 billion on April 4, 2003. Thus, in just a little over two years, George W. Bush has added $732 billion to the national debt. That is just slightly less than the $932 billion that was added to the debt by the first 39 American presidents during the first 192 years of our history!

Wow! That looks like alot. But why is he using absolute numbers and is he discounting for inflation? See, the absolute numbers are not really all that important. Is $100 million dollars alot? What about to Bill Gates? So the amounts we are talking about aren't all that egregious. In fact, as a percentage of GDP Bush isn't even in the top 20, thanks to Calpundit for this information. Why FDR and Truman occupy spots 1-5 on the % of GDP list. So should we be running around screaming hysterically about Bush's tax cuts based on the absolute numbers, absolutely not. One could argue that the cuts are the wrong way to stimulate the economy, but that is something completely different. Sheesh, this guy is not off to a good start.

The rest of that page is a nice long harangue about how awful things are going to be, and even how Bush is going about stimulating the economy the wrong way. What is remarkable is how the initial premise of how complicated the economy is suddenly vanishes from the radar screen.
Steve 10:41 AM

 
Clueless Presidents Lots of people (usually on the left) like to point out how clueless President Bush. Clueless about Iraq, clueless about North Korea, clueless about the economy, clueless about this, that and the other thing. However, it isn't just Bush. I remember back during President Clinton's big push for HillaryCare^TM, it was a townhall meeting and this guy got up and asked the following question (or something like it).

"Mr. President, under your health care proposal my costs would go up by x amount and I can't afford that, with such a program my pizza restuarant business will have to be closed. How can you propose such a policy as it must be the case that I am not the only small business owner in such a situtation."

President Clinton responded with, "Well why can't you just raise the price of your pizza's just $2? I know I'd pay $2 extra for a pizza if it was for such a program...."

President Clinton had some more to the anwer about how it is the "right thing to do", but what struck me was the complete and utter ignorance of economics. Lets look at how markets work. The consumers demand for a good depends on a variety of factors such as prefernces, income, prices, and other parrameters. Firms supply depends on much more concrete items such as capital (plant and equipment), and wages (perhaps a few other items). Now, the new policy will raise costs, this means that some firms will be able to keep operating, but that some firms will be "on the margin", that is the increase in costs will result in shut down. Raising the price wont help this because if the market is competitive no single firm has price setting power. Ultimately, in the end the price may go up, but this wont allow those "marginal" firms to survive because this increase in price will occur because those marginal firms are now gone. It isn't just a simple case of raising the price a couple of bucks. For somebody who ran on, "Its the economy stupid," this was simply amazing to me. Here we have a President who is totally clueless about fundamental economics. I have little doubt that irrespective of the questioners usual political affiliations President Clinton lost a supporter that day with his glib non-answer.
Steve 10:02 AM

 
The Consumer is Sovereign Last week Lynne Kiesling at the Information Problem had a post that touched on a notion that gets introduced to any college student taking introductory microeconomic theory (which is probably a very large percentage of students, but who probably forget it once the class is over since they only care about getting the class done for G.E. requirements), namely the idea of consumer sovereignty.

The idea here is that the consumer is the one ultimately calling the shots, not the suppliers. If you listen to many lefties babbling about anything economic eventually the notion comes up of corporations (particularly eeeevil multi-nationals) that some how control so much. But when you get right down to it the consumer has alot to say about it. Consider the gas crunches during the 70's. Car manufacturers make more money of big gas guzzling luxury cars. Yet, these manufacturers experienced severe financial problems when they were slow to react to changes in consumer preferences. Suddenly the far more efficient Japanese cars looked good, and consumers switched. Eventually Detriot realized what had to be done and started making more fuel efficient cars. Who called the shots there? Detroit or millions of consumers across American? (If you are so stuck on your insipid leftist ideology that you can't bring yourself to say consumers, you are probably in need of professional help).

One example not good enough, how about the Alar^TM scare? Remember that? Consumers suddenly shifted away from apples due to the fear of Alar^TM and the apple industry suffered severe losses that year. Remember Oprah mentioning something about beef on her show once and the resulting lawsuit? Corporations realize that if customers take their money elsewhere they are out of business.

The point is, that firms are not pulling all the strings, and what strings they do pull have only limited effect. In the end, nobody can make you spend your money on something that you don't want to spend it on.

Why do people forget this or not figure it out? I don't know. Perhaps they like the idea of there being somebody other than themselves to blame. When an environmental group comes out about the dolphins getting killed by tuna fishermen, they blame the tuna fishermen. Makes sense, after all if the tuna fishermen are not about to incur additional costs in providing dolphin safe tuna, then the only people left who can make a difference are consumers. Go to them and hope you can convince enough of them to not buy tuna until the tuna fishermen change their behavior. Telling them that they are partly culpable for the death of dolphins and other such ecological problems is just as likely to alienate them as not. So, ignore the consumer's culpability and just point the finger at the eeeevil corporations.

The problem is that some people who don't want to think about this start to get the idea that all the problems are a result of corporations. They aren't. A corporation is out for profits. That is what the shareholders want (they way, you and I...we are shareholders. If you have a 401k, IRA, Roth IRA, etc. you own stock and you are a shareholder). So lower costs, and keep the price as high as possible. Consumers on the other hand want their product and prices to be low as possible. Even the fuzziest headed, birkenstock & hemp wearing, tree hugging environmentalist can probably be caught on occassion wishing some product's price wasn't so high.

Further, asking a firm to do something that has a higher cost unilaterally probably isn't going to work. Lets take a simple example using the dolphin/tuna example. You own a firm that catches and sells canned tuna. Some environmentalists come in and say, "Look, your fishing methods kill X dolphin per year on average, and these creatures are quite intelligent, beautiful, and for all we know are self-aware. This is not a "good thing". We'd like you to implement these new fishing methods that will keep dolphin safe." You look at their suggestion, being a fairly decent person, and you realize it will result in costs going up say 10%. Do you just raise your price 10%? Not if the market is even remotely competitive (and we are talking canned tune...not much of a quality argument here). Raising your price even a little bit can have a serious impact on your sales. Sales go down enough you are out of business (and all your employees are unemployed too don't forget). If you own the company outright you could just eat the extra costs (i.e. reduce profits) if possible, but if you don't the board and shareholders might get upset if you try this. So more often that not the answer is going to be, "Gee, I'd love to help, but there is alot more to this than just the dolphins. I have 500 employees and their families, I have shareholders and their families, and then there are the consumers and their families," or basically the answer is, "No." Of course, if the consumers are swayed by the environmentalists message and stop buying tuna until such dolphin safe measures are put in place, then the changes will be made, and probably pretty quickly.

Quite simply consumers have a great deal of power.
Steve 9:41 AM

Sunday, May 11, 2003

 
U.S. WMD Search Team to Leave Iraq I have to admit that this does not look good. So far the U.S. has not been able to find much of anything suggesting that Hussein had the level of WMDs that the Bush Administration claimed.

Leaders of Task Force 75's diverse staff -- biologists, chemists, arms treaty enforcers, nuclear operators, computer and document experts, and special forces troops -- arrived with high hopes of early success. They said they expected to find what Secretary of State Colin L. Powell described at the U.N. Security Council on Feb. 5 -- hundreds of tons of biological and chemical agents, missiles and rockets to deliver the agents, and evidence of an ongoing program to build a nuclear bomb.

Scores of fruitless missions broke that confidence, many task force members said in interviews.

So, not a pretty picture. Looks like there isn't going to be a "smoking gun" for awhile yet, at best. At worst there will be no substantial finds of WMDs.

Of course, as you read further in the article there are some things that are interesting that many on the left are ignoring.

McPhee saw early in the war that the looters were stripping his targets before he could check them. He cut the planning cycle for new missions -- the time between first notice and launch -- from 96 to 24 hours. "What we found," he said, was that "as the maneuver units hit a target they had to move on, even 24 hours was too slow. By the time we got there, a lot of things were gone."

Short and powerfully built, McPhee has spent his adult life as a combat officer. He calls his soldiers "bubbas" and worries about their mail. "It ain't good" that suspect sites are unprotected, he said, but he refused to criticize fighting units who left evidence unguarded.

"You've got two corps commanders being told, 'Get to Baghdad,' and, oh, by the way, 'When you run across sensitive sites, you have to secure them,' " he said. "Do you secure all those sites, or do you get to Baghdad? You've got limited force structure and you've got 20 missions."

A low point came when looters destroyed what was meant to be McPhee's headquarters in the Iraqi capital. The 101st Airborne Division had used the complex, a munitions factory called the Al Qadisiyah State Establishment, before rolling north to Mosul. When a reporter came calling, looking for Task Force 75, looters were busily stripping it clean. They later set it ablaze.

You don't find stuff like this at Calpundit (which is where I got the link). Right off the bat Kevin dissembles with his title. The search isn't technically over, whatever searching will be done by a new team.

The hunt will continue under a new Iraq Survey Group, which the Bush administration has said is a larger team. But the organizers are drawing down their weapons staffs for lack of work, and adding expertise for other missions.

Overall a very poor job, IMO.

Update I: Henry Hanks has more links on this. By the way, if you aren't periodically stopping by Henry's blog you should consider it. He usually has lots of links on any given issue and you could do alot worse to pick a starting place. At the very least he should be in your favorites list.
Steve 1:29 PM

Friday, May 09, 2003

 
Clinton Hating Thread At DU This is one of those hysterical in an ironic way threads at DU. There are quite a few posts about irrational hatred of President Clinton. What makes it so funny is the irrational hatred described by these DUers is mirrored exactly by the irrational hatred of President Bush by DUers. One newbie who is looking to get tombstoned makes precisely this observation:

Clinton haters were (are) a sad bunch -- irrational, blinded by hatred and incapable of objectivity. Even when he supported policies that they should have embraced as conservatives, they just reflexively opposed him even when it was in their ideological interests to do otherwise. What makes me angry is that there are a lot of progressives whose response to Bush has been no more elevated than that of the Clinton haters.

I've only been on this board since yesterday and I have to say that some of the criticism of Bush is a bit over-the-top. Christ, last night I was reading comments from some people who were insisting that Poppy Bush cowardly leapt from his burning plane over the Pacific and caused the deaths of his fellow fliers during WWII. Jesus. And all this business about Bush knowing about 9/11 in advance is a bit much.

I don't mean to sound presumptuous, especially since I'm new and I'm still feeling my way around, but I'm curiouis if others are uncomfortable with the vitriol directed toward Bush. Am I the only one who sees parallels with the Rethug treatment of Clinton?

No you aren't, but then again the posters at DU think you are probably a Freeper sneaking in to stir up trouble.
Steve 3:04 PM

 
Barney the Blogger Barney, the nom de plum of the guy running MWO Watch Watch Watch Watch, is posting about being banned from commenting on my blog and others. Apparently the guy doesn't even pause and consider that perhaps he is being banned for being a complete ass and swine. For example, the drunkard (Barney Gumble is the drunk on the Simpsons) claims I was using the California electricity market as an example of the free market. Uhhh, okay, if you call criticizing the way the market was set up, pointing out that there was a very real and very serious problem with market power on the part of the generators, and that this is perhaps the best way not to deregulate or restructure a regulated market as some sort of praise, then I guess he is right. This guy isn't banned because he is right, or making very cogent criticisms. He was banned for being a complete idiot, or for just being a troll who knew very well that he was lying and misrepresenting.

So long Barney and be careful not to aspirate your own vomit next time you get drunk.
Steve 12:46 PM

 
Why the Statistics Posts? I was looking at that post immediately below and thinking, "Ha, most people are going to find that dull." Some might even be wondering, "Cripes I had enough of that in college when I had to take that intro stats class, why is he putting this on his site?" Well, I think it is important. Statistics come at us repeatedly through out the day, everyday, either directly or indirectly. How many times do you hear about a poll for this or that on the radio? Or about some study showing this, that or the other thing about your health, diet, car, life? Probably more often than you realize. And underneath all that stuff is statistics. Most of it is from the Frequentist point of view (which is fine), but what most people don't realize is that many of the researchers out there don't really understand it. Not at the level discussed below. Brad DeLong is a case in point (which is why I found his post about revoking Eugene Volokh's tenure for supposedly not understanding the Central Limit Theorem amusing to say the least).

So you get people using procedures and techniques that when you get right down to it, don't really understand. In fact, they probably couldn't tell you much about the procedure even to save their own lives. You see, there are all sorts of software packages out there that will crunch data for any old schlub who has data. Doesn't even matter if the data is good or not. Computers don't care, they just do what they are told to do by their users. So off they go, crunching numbers and reporting the results (usually the results that are deemed "significant"). Then you hear the results and maybe even start changing what you are doing because of it. You change what you do because of some person (or persons) who have gathered data (hopefully good data), and then they crunched this data (hopefully they actually did understand the procedures/techniques they used), and then reported the results if significant. There are several levels here where you have to say, "Okay, I assume they knew what they were doing."

But lets get back to Brad DeLong. He, at least in one regard, did not know what he was doing. Brad teaches at a top school, University of California Berkeley. He is widely published, and many look to him for advice and suggestions. Many assume he is an expert who knows what he is doing. But, in one small way he did make an error. Does this thing happen with others and in ways that are worse? Sure. Here is one example. Stan Glantz wants to use the upper bound of 95% confidence intervals to set policies. Nevermind that we have no reason to assume the upper bound is the better estimate of the parameter. All the 95% confidence interval says is that you are confident that the number is in there...somewhere...maybe, exactly where is not clear, but your pretty confident it is in there.

Robert Matthews also discusses several examples here. One such example are epidemological studies on the efficacy of drugs. It is not unheard of for a drug to look really good in the epidemology studies, but fail to deliver the same levels of success when used more widely. Here is one example that Matthews gives

Again, let us illustrate this through a real-life example. In the early 1990s, the Grampian region early anistreplase trial study (GREAT Group, 1992) generated considerable interest in the medical community, as it seemed to show that heart-attack victims given this clot-busting drug at home had a 50 per cent higher chance of survival than those given the drug once they arrived in hospital. While there were good reasons for expecting that early intervention with the drug would produce some improvement, the size of the claimed benefit surprised many. Nevertheless, frequentist measures of significance appeared to give objective support to the finding: the team found a relative risk (RR) of death for those given the drug early of 0.52 - i.e. a 48 per cent risk reduction - with a 95 per cent CI of (0.23 0.97). As this excludes an RR of 1, this surprising result is also "significant" in frequentist terms, the equivalent P-value being 0.04.

However, as was pointed out shortly after the publication of the GREAT results (Pocock and Spiegelhalter 1992), a considerable amount of prior information existed with which to assess the plausibility of the GREAT finding; for example, a much larger European study involving the same drug pointed to a much smaller benefit. Drawing on this existing knowledge, Pocock and Spiegelhalter carried out a Bayesian re-assessment of the GREAT results; an outline of how such an analysis can be performed is given in the Appendix to this paper. The prior information was captured through a probability distribution which peaked at an RR of 0.83 while giving low probabilities to RRs greater than 1.0 (no benefit) or less than 0.6 (dramatic improvement). When combined with the GREAT data, the resulting ("posterior") probability distribution peaked at an RR of around 0.75, with a 95 per cent Bayesian CI of (0.57 1.0). While still pointing to a more impressive effect than that suggested by previous studies, the GREAT results emerge from the analysis as markedly less impressive than suggested by the frequentist methods.

At this point, it is natural to ask whether this Bayesian analysis really did give a more accurate picture of reality than the frequentist methods. The simple answer is yes. Six years after the publication of the GREAT findings, the overall picture emerging from international studies is that early use of clot-busters like anistreplase does indeed confer extra benefit, with RRs of around 0.75 to 0.8 (Fox, quoted in Matthews 1997). This is only half the improvement suggested by the frequentist analysis of the GREAT data, but in impressive agreement with Pocock and Spiegelhalter’s Bayesian analysis.

Here is another problem that Matthews points to

To see the dangers inherent in this fallacy, suppose a patient walks into a doctor’s surgery covered with spots. The doctor knows that the probability of getting spots given a measles infection is very close to certainty, i.e. Prob(spots | measles) ~ 1. However, it clearly does not follow that the probability that the patient really has got measles is also close to 1, i.e. that Prob(measles | spots) ~ 1: there is a vast number of other diseases apart from measles that produce spots. Deciding which the patient has will involve taking into account other sources of information, such as whether there is chicken pox in the family, and whether the patient has recently travelled abroad.

Clearly, mistaking Prob(spots | measles) for Prob(measles | spots) could lead to a doctor being struck off. Yet the standard methods of statistical inference can and do prompt working scientists to fall into precisely the same trap: P-values are all too easily taken to be identical to Prob(null hypothesis | data), so that a low P-value is taken to imply that the probability that chance alone explains the data is similarly low. There is no simple relationship between P-values and the probability working scientists actually want, and as I shall show shortly, confusing the two can and does lead to meaningless fluke results being regarded as "significant".

So understanding precisely what statistics techniques and procedures are is important for the researcher, IMO. It doesn't mean you (the layperson) has to have advanced knowledge, but just realize that impressive statistics may not really be all that impressive. Keep in mind that while these procedures usually work, they can and do fail. Also, remember that people might apply the wrong procedure, do it incorrectly, etc.
Steve 10:03 AM

 
More Statistics Blogging Considering the good responses to my last post on statistics I thought I'd try another. Two of the main schools of thought in statistics are the Bayesian and the Frequentist schools. They are in many ways very different, almost diametrically opposed. One area is precision. The Frequentist school takes the view of initial precision while the Bayesian view focuses on final precision.

Initial precision is where the focus is on the accuracy of precedures prior to any sample data being gathered, let alone analyzed. That is, over many, many trials a procedure set up with initial precision in mind will perform well most of the time. This idea is demonstrated in confidence intervals, which incidentally most people have a heck of a time understanding correctly (admittedly the logic is really convoluted and arriving at the wrong conclusion is actually not that hard). This view fits nicely with the Frequentist definition of probability that I discussed here.

Final precision on the other hand has to do with accuracy of the estimator (or procedure) after the sample data is collected. This view lends itself quit naturally to the Subjective or Logical definitions or probability. Another, aspect of this approach is that the researcher does not have to worry about data that might have been observed, but wasn't.

One aspect of this distinction is that sometimes the procedures developed under the initial precision philosophy often look really whacky. You get confidence intervals that are too big, for example. Another problem is that confidence intervals convey a level of accuracy that is not warranted--e.g. if the sample is uninformative.

Now as to the definition of confidence intervals. First lets start with the procedure itself--i.e., the probability interval:

P(W₁ < g(θ) < W₂) = 1 - α.

Where W_i are two statistics. This is indeed a probability interval. Over many, many different samples this interval will contain g((θ) with probability 1 - α. However, W_i ≡ ω_i(Y₁, Y₂,...,Y_T), that is no actual data has been observed yet. If we replace W_i with the observed values w_i (i.e. we have observed the data and w_i are now fixed) then you now have a confidence interval:

C(w₁ < g(θ) < w₂) = 1 - α.

Note that nothing is random anymore. Given the frequentist view of parameters, g(θ) is fixed. Given the nature of frequentist estimators w_i are also fixed. Everything is fixed. So how can you say that such an interval has a probability associated with it? You can't.

Think of it this way, suppose you are holding a fair 6 sided die. Now if you throw it, and it stops moving, what is the probability that the top side is showing a number other than what it is showing? You should say zero. What is the probability that it is showing the number it is showing? You should say 1. Confidence intervals are basically the same, but the problem is you don't know where g(θ) is so you don't know if it is in any given interval or not. But g(θ) is fixed and so is any given interval you have constructed. There is no probability, except in the degenerate (0,1) way, that such an interval contains g(θ). All you are is confident in the procedure that you used to construct the confidence interval. That procedure works, 95% of the time. But in any given instance, when all is said and done (i.e., you have collected the data, crunched the numbers and have one of these intervals) you don't know if it worked or not. You are simply confident that it worked.

Granted this is a subtle point, but it is crucial. Frequentists don't see things like g(θ) has being random. The parameters you are interested in have an objective existence and are fixed. To use an example from economics, the marginal propensity to consume at any given time is fixed. There is no randomness here, it has one single value. The problem is you can't observe it directly so you use statistics to try and infer its value. These inferences either worked (i.e. the confidence interval contains the actual value) or it didn't. Now a Bayesian (or Subjectivist) would say that g(θ) is not fixed (and may not even have an objective existence, after all we are talking about an economic model here, not reality). So we can talk about it in a probabilistic manner. Now the Bayesian is going to calaculate his interval estimate directly from his posterior probability distribution. Since it is a probability distribution making probabilistic statements such as, "this interval contains the parameter with 95% probability..." is correct. Oddly enough, in most examples the numbers associated with the intervals are identical, but it is the interpretation of those numbers that are different.
Steve 8:59 AM