He talks about the Fed and Treasury here. Bair would be a superb choice.
Monday, November 12, 2012
Sunday, November 11, 2012
Coastlines and votes
President Obama did better on the coasts and the Great Lakes states than elsewhere. I thought it would be fun to plot coastline/shoreline miles by state against Obama vote percentage in 2012. Here is what I got:
The data on coastline/shoreline come from http://www.michigan.gov/deq/0,4561,7-135-3313_3677-15959--,00.html and The Statistical Abstract of the United States, Table 364. Vote totals come from Dave Leip's Atlas of Presidential Elections. The correlation is .22, so not huge, but not nothing either.
If one adds Alaska, with its 6000+ miles of coastline and support of Romney, the correlation goes to zero.
If one adds Alaska, with its 6000+ miles of coastline and support of Romney, the correlation goes to zero.
Who is doing the shopping?
I was listening to a retailer last week discuss the interaction between housewives and grocery stores. The tone of his remarks suggested that he thought housewives were grocery stores' principal customers.
This didn't seem right to me, so I tabulated the Family Type and Employment variable in the 2006-2010 sample of the American Community Sample (having a laptop with flash memory is pretty awesome--it allowed me to do it in about five minutes).
Here are the results:
Household Type # of Households % 0f Households
Married Couple, both in labor force| 22,309,285 41.16
Married Couple, only man in labor force 8,792,744 16.22
Married Couple, only woman in labor force 2,933,758 5.41
Married Couple, neither in labor force 6,473,742 11.94
Male only, in labor force 2,848,830 5.26
Male only, not in labor force 720,066 1.33
Female only, in labor force 7,230,003 13.34
Female only, not in labor force 2,890,776 5.33
This didn't seem right to me, so I tabulated the Family Type and Employment variable in the 2006-2010 sample of the American Community Sample (having a laptop with flash memory is pretty awesome--it allowed me to do it in about five minutes).
Here are the results:
Household Type # of Households % 0f Households
Married Couple, both in labor force| 22,309,285 41.16
Married Couple, only man in labor force 8,792,744 16.22
Married Couple, only woman in labor force 2,933,758 5.41
Married Couple, neither in labor force 6,473,742 11.94
Male only, in labor force 2,848,830 5.26
Male only, not in labor force 720,066 1.33
Female only, in labor force 7,230,003 13.34
Female only, not in labor force 2,890,776 5.33
I am guessing that most retailers know and understand the implications of a country where only 16 percent of households have housewives (in the traditional sense of the word), but perhaps they are not.
Saturday, November 10, 2012
I am not thrilled...
..that President Obama has made Tim Geithner the point person for budget negotiations. I very much hope he doesn't negotiate away the President's strong position.
Friday, November 9, 2012
Monday, November 5, 2012
Another thought on Nate Silver
If poll errors are randomly distributed across states, then Obama has it won. But if they are not, Romney still had a chance. If how voters break is one state relative to polls is correlated with how they break in others, then if they break toward Obama, it makes no difference to the ultimate win-lose outcome; but if they break toward Romney, it does make a difference.
So statistical independence in errors across polls is good for Obama; correlation is bad for him.
So statistical independence in errors across polls is good for Obama; correlation is bad for him.
Subscribe to:
Posts (Atom)






































