Themes & Topics
Amid Rising Bond
Yields, Six Experts Look Ahead
On February 4,
1996, the 30-year Treasury bond yield increased to 40%,
its highest level in four months. The WSJ asked
some "experts" about their forecast of yield at
year-end. Below are their reported estimates:
yield at year-end
Straszheim, chief economist
head of quantitative research
What does this suggest to you?
Obviously, interest rate forecasting is very
difficult and seems to depend heavily on the specific
assumptions that the forecaster makes.
If you were to form your own "view" about
yields, what might you do with these numbers?
A: A simple
answer might be take the average of these numbers.
However, there might be a better solution. Instead of
using a simple average, you might want to use a
"weighted" average, i.e., assigning
different "credibilities" to the forecasts.
But how might one come up with a weighting scheme? At
least on an intuitive level, you might give more
relative weight to the one who historically had
"better" forecasts, or the one who has been
more accurate in the recent past. Technically, you
can use statistical techniques to put more rigor into
these intuitive explanations.
Source: WSJ, 2/26/96, p.
Q: Why did God create
order to make weather forecasters look good.
have forecasted 9 out of the last 5 recessions.
We have 2
classes of forecasters: Those who don't know . . . and
those who don't know they don't know.