Multiple comparisons

Wikipedia describes this problem as the following:

In statistics, the multiple comparisons or multiple testing problem occurs when one considers a set of statistical inferences simultaneously.^[1] Errors in inference, including confidence intervals that fail to include their corresponding population parameters or hypothesis tests that incorrectly reject the null hypothesis are more likely to occur when one considers the set as a whole. Several statistical techniques have been developed to prevent this from happening, allowing significance levels for single and multiple comparisons to be directly compared. These techniques generally require a stronger level of evidence to be observed in order for an individual comparison to be deemed "significant", so as to compensate for the number of inferences being made.

Some examples (again from Wikipedia):

• Suppose the treatment is a new way of teaching writing to students, and the control is the standard way of teaching writing. Students in the two groups can be compared in terms of grammar, spelling, organization, content, and so on. As more attributes are compared, it becomes more likely that the treatment and control groups will appear to differ on at least one attribute.

• Suppose we consider the efficacy of a drug in terms of the reduction of any one of a number of disease symptoms. As more symptoms are considered, it becomes more likely that the drug will appear to be an improvement over existing drugs in terms of at least one symptom.

• Suppose we consider the safety of a drug in terms of the occurrences of different types of side effects. As more types of side effects are considered, it becomes more likely that the new drug will appear to be less safe than existing drugs in terms of at least one side effect.

The question I have is the following: What if I refuse to test for some of the additional attributes/symptoms or hypotheses? Does this make the multiple comparison problem go away? If I ignore the fact that there are possibly other hypotheses out there which I do not specifically test for does this make my results more valid in the sense that I do not have to adjust for the confidence interval or level of significance? What if all studies did this - just focus on the hypothesis of interest and ignore all other testable hypothesis within their model or experiment?

Do recessions begin with layoffs?

According to Robert Hall (Restat, 2005), Employment Efficiency and Sticky Wages: Evidence from Flows in the Labor Market:

... recessions do not begin with a burst of layoffs. Unemployment rises because jobs are hard to find, not because an unusual number of people are thrown into unemployment.

What does the data say? From FRED, the monthly total layoffs and discharges show that the current recession did not begin with layoffs:


This assumes among other things that the NBER dating committee correctly pinpointed the date of the recession. Could this also be a result of seasonal adjustment? The following is unadjusted totals:


And unadjusted rates:



The plots show that layoffs occur regularly with clear cycles. Layoffs can be higher or lower in or out of recessions.

Unconvinced is Shigeru Fujita. The following is Figure 4 of his paper, and one would expect that in order to say that Hall is incorrect, the job loss rate should show a spike just prior to the beginning of the recession and furthermore this spike should be higher than all the job loss rates prior to the recession (from the trough of the previous recession).


Perhaps this is too much to ask of the data, and as much as I hate to admit it, perhaps Hall is right. I began this post thinking I could prove him wrong.

What makes a book a children's book

What makes a book a children's book (as opposed to a young adult book)? I've been reading Bridge to Terabitha, Wrinkle in Time, Whale Rider, Tuck Everlasting, Lois Lowry (The Giver, Gathering Blue, The Messenger and Number the Stars) and Philip Pullman. Length and language are certainly two criteria although I did not discern any dumbing down of the language in any of these books. It's hard to argue that Philip Pullman's Dark Materials trilogy (running almost 1000 pages in total) or even Harry Potter (which I have yet to get to) have been shortened for children. In fact the themes that Philip Pullman deals with – the original sin, the church and theology are not what can be considered topics that children think about and in the degree of violence in his books were almost too much (for me). And even though I don't consider myself religious, the anti-Church message was a little over the top. In fact, none of the afore mentioned books deal with what I would consider children's themes e.g. horses, pets, fairies, or anything in this vein. These books have the common thread that the main characters in the books are children – but Orson Scott Card's Ender's Game also have children as the main characters but is not considered a children's book. Is it all marketing? Jenny Downham's Before I Die is definitely a young adult book since it contains sex – and includes the incredible feat of a first person narrative who dies at the end. But what about the Twilight series – is it merely because the characters are young adults versus for instance Ann Rice's Vampire series? Again, is it all in marketing?

Supporting the arts

It seems almost universal that no matter what the culture, we love to celebrate the successes of artists, performers and sportsmen. We have been to several school performances this school year and there is no doubt that there is a lot of talent – yet when is their talent good enough that we should encourage and cultivate it? There is also no doubt that talent by itself is not enough for commercial or even limited artistic success. Do we sow the seeds of desire in these young men and women to strike out into the artistic world when we clap and shout 'Bravo!' or 'Author, author!'? And if so, are we responsible for helping them find their way whatever that may be – or should we tamp down our enthusiasm and clap politely and whisper 'Don't quit your day job'.

For while the returns to college may be an expensive gamble, so is the returns to artistic and sporting endeavors. Just how much should we as parents encourage any kind of artistic or sporting career or be responsible for their chosen paths?

Reading Posner on the financial crisis

The books are The Crisis of Capitalist Democracy, which preceded A Failure of Capitalism: The Crisis of '08 and the Descent into Depression. Here is Robert Solow's reading of the latter. Both books have a feel of being written in a hurry and as such sounds disjointed in its narrative. The first book is essentially an impression formed as the crisis evolved and is light on prescription and heavier on description though not quite as moving as other books on the crisis. I detected a shift from the earlier book in terms of causes of the financial crisis. The first book concludes as most onlookers did at the time that excessive leverage was one of the prime causes of the crisis. While not blaming leverage per se, Posner discusses this as well as other factors that may have brought about the crisis. In the latter book, he argues that the causes of the crisis stem from low interest rates which caused a bubble to form, as well as lax supervision in the belief that markets would be self-regulating since the ideology of efficient markets had taken hold.

At least, this is my reading of it and I am surprised that Solow did not touch on this aspect at all. (See also a review from the NYT which has a similar interpretation as this.) Coming from Posner who is a well known conservative and from the bedrock of free markets of the University of Chicago, I found this to be quite a turn around and in parts of the latter book he admits this freely. All the other 'causes' such as greed, CEO pay, excessive leverage, creation of new derivatives can be explained rationally from these two factors. (See also here.) In this sense, the second book is better than the first and what is also refreshing is his admission that he really does not see anything from current or pending legislative reforms that would avoid another crisis. While he admits there are many good ideas out there such as counter-cyclical leverage requirements, he is not optimistic that these will come to pass.

In all, the second book is better than the first though it could have been a better book had it been more tightly edited. In some passages, he sounds as though he is rambling and going off on a tangent.

What people do where

This is something that strikes me whenever I'm on vacation and having discovered On the Map, I can look anything up:
1. Kitty Hawk: 28% Retail, 13% Health and 12% Food and Accomodations (2,524 jobs in 2009) and almost 50 percent of the jobs do not require a college degree (although about 25% of the jobs do no list education requirement)
2. Nags Head: 21% Health, 16% Food and Accomodations, 14% Retail (3,106 jobs in 2009) and like Kitty Hawk, almost 50% don't require a college degree - with the same degree of missingness in the data.
3. Kill Devil Hills: 17% Health, 16% Retail, 15% Food and Accomodations (3,076 jobs in 2009) and like above, most jobs do not require a college degree.

On The Map also allows to do an inflow-outflow analysis of Nags Head and Kitty Hawk. Drawing a polygon from Kitty Hawk (around the Wright Memorial Bridge) down beyond Hatteras shows that the number of people living in and working in this area have stayed steady at around 50% between 2007-2009. This says to me that the people aren't being priced out of the neighborhood just yet.

OBX

Spent most of this week at the Outer Banks. Those who live there call themselves (some of them anyway), Outer Bankers. (There are some who think it is all right to be bankers.) We rented a house in Nags Head found via VRBO. The house was between Highway 158 and 12 and fortunately, since it was early in the season, it was still relatively quiet. The only real disturbance in the morning was the construction nearby on Jeanette's Pier.

I found it remarkable that despite being less than half a mile on the ocean on one side and the sound on the other, it was very easy to forget that we were by the sea. Nags Head and most of the strip from Kitty Hawk is laid out in a grid with commercial strip malls on either side on 158. On Highway 12, there are houses, restaurants and motels/hotels on either side, making it convenient since we were close by to several restaurants (within 5 minutes walk) as well as a convenience store and a 7-11.

All the homes have balconies, some more than one - like outstretched arms embracing the sea, as if saying, "Take me home." It will not be surprising to me to read that life on earth began from the sea.

I also found it contradictory that when on vacation I like isolation and yet be close to all the trappings of suburbia at the same time. It would have been nice if it were possible but the scale of development of the Outer Banks precludes that. We drove as far north as Corolla, where houses like this were the norm. The northern part of the Outer Banks from Kitty Hawk feels more affluent as evidenced by the size of the homes and the lack of public beach access. Our last night was spent at the Hilton Garden Inn in Kitty Hawk - oceanfront - we could hear the waves from the room. The bed was too soft for me and the space was a letdown after a house but it was otherwise a comfortable stay.

The weather was cool but for the most part good and the food at FatBoyz, Owens and Outer Banks Brewing were delicious. I'm not planning on going back - it was one of the places on my list that I can now cross off. It's too bad we did not have the inclination to drive down to Cape Hatteras.