Re: Letter form Bob Solow

[pdavidso <pdavidso@xxxxxxx>]

>===== Original Message From "Lonnie K. Stevans" <acslks@xxxxxxxxxxxxxxxxx>
=====
>Sorry, ignore my previous posting:
>
>Sounds like a Type II error to me.  It is certainly plausible to observe the
presence of a unit
>root when the alternative hypothesis of no unit root is true, e.g., if the
probability of a Type
>II error was .10, then by performing 100 Monte Carlo simulations of white
noise, you will
>wind up making the "mistake" of accepting the null hypothesis of a unit root
ten times.
>Yaglom just happened to observe one of those outcomes and it would seem to me
that
>he should have repeated the experiment a number of times.
>

between Yaglom and graduate students replicating the experiment it was done
more than 100 times -- with the original columns showing unit root (would you
expect anything welse from two columns of random numbers?)

And the n-th difference sowing statistical significance.

So your argument about a type II error in all these cases just does not hold
water.

Paul

Paul Davidson
Editor, Journal of Post Keynesian Economics
University of Tennessee
SMC 523
Knoxville, Tennessee 37996-0550
phone # (865)974-4221; fax #(561)737-8262;
email pdavidson@xxxxxxx
http://econ.bus.utk.edu/davidsonextra/Davidson.html