Mr. Osburn,
In a recent colleague review of a paper regarding the Durango shutdown you
asked for additional information on the smoothing factor or alpha as
referenced by Cleveland (1994). The standard reference that I use for
Loess is Cleveland, W.S., and Devlin, S.J., (1988) Locally-weighted
Regression: An Approach to Regression Analysis by Local Fitting. J. Am.
Statist. Assoc., Vol. 83, pp 596-610. In this there is no mention of a
smoothing factor and I want to avoid confusion (As you know Cleveland was
the world's worst for changing terms--witness the confusion between LOWESS
and LOESS). No one in my office or the regional office has a copy of this,
and so I would need to get a copy on interlibrary loan. This would take a
couple of weeks. If you could fax the pertinent portion of the article to
me at 770-903-9199, I could address your technical concerns in a more
timely manner. Thanks so much for spending the time to review the article.
Sincerely,
Chris
Chris Leeth
U.S. Geological Survey (WRD)
Atlanta, GA
|