I am going through this with a co-author right now. His vision was to use raw data put out by the NHL. My reality was that this data was over 1,000 observations on over 1,200 games. I went with the script:
Now that it’s done (thanks to Trevor MacDonald) I know I am way out on the right of this chart.
The conceptual basis of classical hypothesis testing is useful in a wide variety of areas.* Pity most students get too caught up in following the recipes for constructing test statistics to realize this.
A Type I error is foreclosing against a good borrower. A Type II error is letting a bad borrower avoid foreclosure. I will readily grant that a Type I error is worse than a Type II error, so we should tolerate some of the latter in order to avoid the former. However, I contend that we have let this bias get completely out of hand, resulting in a huge pileup of Type II errors with catastrophic effects on the housing market.
This is in regard to a Maryland family that has gamed the system and not made a payment on the house they continue to live in for over 5 years.
* A Type I error is when you make the mistake of rejecting something that is true. A Type II error is when you make the mistake of not rejecting something that is false. For example, a Type I error is convicting someone who is innocent, while a Type II error is letting someone go who is guilty.
The probability of being hit by lightning is low, but the conditional probability* may be much higher.
As a Utah resident, I can tell you why the death rate is higher in these dry western states.
First, lightning often strikes exposed hilltops, mountaintops, cliffs, and so on. Utah has a lot of those. Also, they’re scenic and accessible, so we have a lot of people out on them.
Second is virga. Easterners don’t usually even know what this is. Virga is when a storm drops rainfall and it evaporates before hitting the ground. We have a lot of virga in Utah. The thing is with virga, even though you’re dry, that’s a real live storm up there.
Third is visibility. Where I live, I can see for 60 miles on a bad day. It’s routine for me to be able to see multiple storms, separated by clear sky. And, it can be really easy to stop worrying about lightning when you see lightning, and count to over 100 before you hear thunder. Lightning routinely strikes as much as 10 miles from a storm, and up to 50 miles (who knew!). So, in Utah, not only can you see multiple storms, you can be hit by them too.
Fourth, there is a lot more lightning in sub-tropical areas, but there’s also a lot more rain. I lived in Louisiana for almost a decade: if there is a thunderstorm outside, you simply don’t go outside because it is like walking into a running bathroom shower with your clothes on. You wouldn’t jump into a pond fully clothed, and in Louisiana you don’t go out in the rain if you can avoid it because sometimes you’ll get that wet in a minute or two. This doesn’t happen out west, so people are outside more when there’s lightning.
* Probability is the chance of something happening. Conditional probability is the chance of something happening when you know some other information. Ben Franklin wasn’t dumb because the probability of being hit by lightning was high, but because the conditional probability of being hit while flying a kite is much higher.
† Do you ever wonder if some people’s teenage years would have been easier if they changed their names?
This is like not using batting average (at all) to compare baseball players because there is much more to the game.
Repeating the paraphrase of Orwell “Some ideas are so stupid that only intellectuals believe them.”*
Here’s Bryan’s result:
… If you mention ability bias, however, labor economists will quickly point you to a massive literature that supposedly debunks it.
But if you pay close attention, there's a bizarre omission. Despite their mighty debunking efforts, labor economists almost never test for ability bias in the most obvious way: Measure ability, then re-estimate the return to education after controlling for measured ability. For example, you could measure IQ, then estimate the return to education after controlling for IQ.
When I ask labor economists about their omission, they have a puzzling response: "IQ is a very incomplete measure of ability." True enough. But the right lesson to draw is that controlling for IQ provides a lower bound for the severity of ability bias. After all, if the estimated return to education falls sharply after controlling for just one measure of ability, imagine how much it might fall after controlling for measures of all ability.
What happens to the return to education after controlling for IQ? I've done the statistics myself on the NLSY, and found that the estimated return to education falls by about 40% …
So, the reported returns to education — how much extra one gets paid for more schooling — are almost double the corrected estimates because they’re ignoring the basic observation that smart people are likely to get paid more anyway.
The flip side of this is that the value of teachers and schools is overstated by ignoring the quality of the students put into the system.
Gee … do’ya think there’s any grant funding out there interested in establishing and promoting that result?
Statistically, Bryan is pointing out that conventional parameter estimates are biased upward, but he’s missing a second problem. The big three problems in statistics are bias, consistency, and efficiency — and the conventional estimates are not just biased, they’re also inefficient.
Inefficiency is a lot more subtle than bias. Frankly, I don’t think I understood it until about 8 years into my academic career when I had to explain in seminars to finance Ph.D.’s why they should listen to an economics Ph.D. about the problem with one of their techniques.
In short, an inefficient estimator is one that’s just dumb: like forecasting the weather without looking at the sky. There’s a lot more to it than that, but it means that you’re not doing something which might be helpful.
One way to think about this is that bias is about parameter estimates that are off in one direction or the other, while efficiency is about standard error estimates that are off in one direction or the other.
In practice, this suggests that not only are the typical estimates of the returns to education biased upwards, but they probably reported to be a lot more precise than they actually are. That is, Bryan is pointing out that the parameter estimates are biased upward, and I’m pointing out that the standard error estimates are probably biased downwards.
Bias and inefficiency also have distributional consequences for the conventional results. Because those results are based on leaving something out, they suggest that the things that are left in are more important than they would be in a better model. The thing is, you don’t know which of your variables is more or less seriously affected until you run the better model. Maybe you’re lucky, and the variable of interest is the one that’s least affected … and maybe you’re not.
With the development of internet technology, work at home jobs are increasing in the market. Also setting up small business online with ones own bank savings can provide excellent work at home opportunities. Apart from savings, banks offer0 credit card to cater to short term finance needs. Partial tax payments like tax credits are also available to promote online businesses. Market now offers several alternatives to traditional credit card debt which are helpful to work at home businesses.