Posts Tagged With: Uncategorized

The Role of Prediction Markets in Forecasting

Robin Hanson summarizes the literature, in case you weren't convinced:
Decades of research on financial markets have shown that it is hard to find biases in market estimates. The few direct comparisons made so far have found markets to be at least as accurate as other institutions. Orange Juice futures improve on National Weather Service forecasts (Roll, 1984), horse race markets beat horse race experts (Figlewski, 1979), Oscar markets beat columnist forecasts (Pennock, Giles, & Nielsen, 2001), gas demand markets beat gas demand experts (Spencer, 2004), stock markets beat the official NASA panel at fingering the guilty company in the Challenger accident (Maloney & Mulherin, 2003), markets on economic statistics beat surveys of professional forecasters (Grkaynak & Wolfers, 2006), election markets beat national opinion polls (Berg, Nelson, & Rietz, 2001), and corporate sales markets beat official corporate forecasts (Chen & Plott, 1998)

Liked what you read? I am available for hire.

Tortured logic

If an interrogation technique is painful enough to make a person divulge information that they would not have otherwise divulged, in order to make the pain stop, it's torture. It's that simple. You can't use pain to gain a confession. There are studies of evidence showing that tortured prisoners will confess to anything and everything to avoid further pain. If you don't believe it, read 1984 again. If a technique is so soft (as "walling" supposedly is) that we can describe its effects on the prisoner as "not torture," then it probably wouldn't make anyone reveal anything. But if that assumption is correct, then why were the interrogators using the technique? The only possible argument is that the information we received from torturing prisoners prevented attacks on US, and this benefit outweighs the harm to our reputation, our flaunting of the Geneva Convention, the reciprocal pain that American POWs can expect from now on when they're captured, and the droves of Arabs who want to join terrorist groups to get back at us for torturing their fellow citizens. But we've been given no evidence that the information gained from torture foiled any specific plots. In any event, the marginal benefit from the 183rd waterboarding of a single prisoner is low. We tortured prisoners. Someone high up in the Bush administration either authorized this torture, knew about it and allowed it to proceed, or had no idea, which given this administration, is possible, but sad.

Liked what you read? I am available for hire.

PP of the day

“Here’s how to know if you have the makeup to be an investor. How would you handle the following situation? Let’s say you own a Procter & Gamble in your portfolio and the stock price goes down by half. Do you like it better? If it falls in half, do you reinvest dividends? Do you take cash out of savings to buy more? If you have the confidence to do that, then you’re an investor. If you don’t, you’re not an investor, you’re a speculator, and you shouldn’t be in the stock market in the first place.”
From Jeffrey Goldberg in this month's Atlantic. Goldberg's panicking and selling at the bottom, giving the article a view of the market from a speculator's perspective. The quotes from Robert Soros are particularly telling (and remind me  of my post from last week). I'd recommend Less Antman, and fee-based planners in general, to Jeff.

Liked what you read? I am available for hire.

Mental Health Break

Video of the Big Dog robot we watched today in class: The way it recovers from slipping on the ice and being kicked is incredible. It's carrying a 300-pound payload on its back. Here's an early version of the same project: Can you spot the difference?

Liked what you read? I am available for hire.

Most investment advice is terrible

Investing There are two schools of thought on investing as a college student. One says that because your current income is low and your future income is high, spend money now because you're going to make more later. The other says that the earlier you start investing, the higher your return will be in the future, because with compounding and exponential growth, you make the majority of the return in the last periods of investment. I take a middle road; I save when I can but I spend a large chunk of my money. The preponderance of information on the internet likens investing to gambling. Get this straight: If someone has a proven, mathematically rigorous method for beating the stock market, they will not tell you about it. It would be like Coke giving away its formula or KFC giving up its chicken recipe. If you have a good tip, and you tell other people, they'll bid up the price until it's no longer a valuable tip. The repercussion is simple: all financial advice telling you about ways you can beat the market is complete and utter bullshit. All of the advice sounds really good; we love gambling, we love new ways to gamble and we love thinking that we have a foolproof strategy. But the only people that make money off of this advice are the givers and the market makers. I'll explain why in a little bit. I make a small exception for people who are giving advice based on how they have their own money invested; at least if they're wrong, they're losing money too. For example, Warren Buffett's letter to shareholders, where he discusses how his company is invested in light of the current market climate. I got the following in my email box recently, from an online trading company:
Here are a few reasons we think you’ll find forex trading interesting: Large and Liquid Market Forex is the most traded market in the world, with a daily volume exceeding $3.2 trillion, according to the Bank for International Settlements (BIS).  Alternative Asset Class Zecco Forex provides access to another asset class where you can make trades independent of overall equity market direction. So if the Dow is plummeting, the dollar could be rallying. And if the dollar is plummeting against the Euro, you might decide to jump on the Euro and ride that wave.  Leverage You can get up to 200 to 1 leverage trading foreign currencies. That means $500 cash can purchase up to $100,000 in currencies (my note: and a 2 percent loss will not only wipe out your $500 investment, but leave you owing a further $1500!). Now before you go investing your kid's college funds in currencies, you should recognize that a high degree of leverage can lead to large losses as well as gains. As with any other investment, you should only risk what you can afford to lose.  Technical We believe forex trading will be very attractive to active and technical traders. With a multitude of trading opportunities, and a full array of technical strategies, we believe you will find this market worth trading. And because we’re Zecco, you’ll get some of the tightest bid-ask spreads in the business. 
Every transaction has two sides; a buyer and a seller - someone who thinks the stock's going up and someone who thinks it's going down. If you don't think you know more than the other party (an institutional investor, a hedge fund trader, a math Ph.D in the basement at Goldman Sachs), don't trade. Note how much the email compares this to gambling - they cover every possible investor, from the guy who wants to "ride that Euro wave" to the one with "a full array of technical strategies."   A simplified story of how to invest money: Let's say you're at the horse races for the day, at a special horse track called "Dow Jones Fields." Normally, a horse track will post odds, collect everyone's money, remove a 15% cut and then pay out the remaining 85% to the winners - this is called parimutuel betting, because your winnings are funded by your fellow man's loss. Because this is a special track, the return varies - some races, the owners remove 20% of the money pool, some races they put in 30%, and over every race at the track we know that on average, at the end of the day, the owners (Mr. and Mrs. Profits) pitch in four or five percent to the total pool, so there's more money out than went in. Given that this is a racing track, there are lots of individual horses, each with a different record, a different jockey, different starting gate, different odds etc. In addition each horse at this track has qualities called "Beta" and "P/E Ratio" and "Projected Analyst Earnings." We have a pocket full of cash and three possibilities: we can earn below 4-5% return, we can earn exactly 4-5% return, or we can try to beat the market. Three levels up, you can see a bunch of guys in suits and sunglasses making bets. They're trained in horse betting, and make their living making money at the tracks. They're your competition. Now, the surefire way to lose a lot of money is to look at which horses have been winning, figure out who their jockeys are, what their betas are, and what their odds are, and then bet those horses in future races. As humans, we're more than capable of making patterns out of random data. This is exactly what we do when we gamble: in roulette, bet on a "hot" number, or whatever. We could try to bet on the "hot" horses. And we might get lucky and win but odds are we're going to lose, or at least earn less than 4-5%, because of the guys in the suits at the top who know everything and make bets for a living. The best thing to do is to bet every single horse in every single race. This way you guarantee you'll get the market return. Now in an individual race you may be way up or way down depending on how big the pool is. But we more or less know in the long run things are going to be positive.

Liked what you read? I am available for hire.

So-called “prep schools” for elite young athletes make sense

I read this Sports Illustrated article about Findlay Prep, the "school" set up by a UNLV booster for top young players. They live together in a house with plasma-screen TV's, equipment, food and furniture all paid for by the booster. Spoiling young athletes seems ridiculous at first glance, because it seems like it's circumventing the rules that everyone else plays by, but the spoiling is a consequence of a system where the players are not allowed to be paid directly until they graduate or leave college. It's silly that we force top athletes to go through the farce of high school and college classes, with the idea that they are "students first," as all those silly NCAA commercials remind us, who happen to be extremely talented at sports. In sports the normal economic relationship between Europe and the USA is reversed; European players enjoy contractual freedom, while the American system limits players freedom to earn market wages and play for the teams that they want to. In Europe, the best young players live full time at sports academies, where they take some classes but spend most of their time training and practicing. Top clubs often sign and recruit players at young ages; compensation and contract length for players under the age of 16 is limited. Manchester United recently made headlines for signing a 9-year old, but it's pretty common practice in England - the United spokesman said "the club signs about 40 players of Davis's age every year and, as is standard, will decide annually whether to renew his contract or release him." Furthermore, there's no such thing as a draft - players sign contracts with teams. Contracts can be bought or sold but player trades are rare (I've wondered why this system hasn't caught on in the US - paying cash for players means you can express their value more directly, rather than through the players on your team). Universities field teams, but their players are drawn from the student body - "recruiting" is a little outdated when pro teams can and do pay wages to players directly. But instead of actually compensating young players for their current and future value to the teams they play for, colleges go through an elaborate courting process, recruiting, and pro teams gather every June to take turns drafting players - for the first time they'll be able to earn money playing basketball, but they get no say in where they can play. So I believe the academy system makes more sense - top players can focus on basketball, and universities can admit more qualified students.

Liked what you read? I am available for hire.

What can we predict about the future?

In no particular order: 1) Government intervention tends to have negative consequences for society as a whole. 2) The best predictor of what will happen next time is often what happened last time. 3) Heuristics (knowing too much) tend to make us worse predictors. The classic example of this is the study where German students were asked whether San Diego or San Antonio is bigger. I'm not convinced about that one statistic proving a general rule, but I'm convinced about the rule. 4) Over time, we'll become less mad about people who have aggrieved us. We'll also recover from the loss of loved ones. 5) We will change our minds; conclusions ("leverage is a bad idea") will seem obvious, even though they weren't at the time. 6) One of my favorite movie scenes is in American Beauty, when Ricky asks his dad, "Whats going on in the world, Dad?" and Chris Cooper puts down his newspaper and says "Son, this country's going straight to hell." I feel like people have always said this and will always continue to say it. In my worst liberal days in high school I believed it. The world tends to do all right though. 7) If someone practices a particular skill, with persistence and deliberate feedback, after a while they'll improve.

Liked what you read? I am available for hire.