[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
RE: [riskanal] RE: If you do Science, use the Scientific Method!
Dear Frank,
You are right on! For years now, Joe Alvarez and I have faulted
the LNT proponents in papers, countless presentations, and many
discussions, for treating the dose or exposure as non-stochastic
quantities. That is simply and outrageously wrong! We all know
that doses or exposures are stochastic quantities with often
quite large errors. We pointed that fact out in an old paper in
the Proceedings of the First Dixy Lee Ray Symposium in 1994:
Seiler, F.A.; Alvarez, J.L. "The definition of a minimum
significant risk. Technology 331A: 83-95; 1994,
and more recently in the paper:
Seiler, F.A., and J.L. Alvarez, "Is the 'Ecological Fallacy' a
Fallacy?" Hum. Ecol. Risk Assess., 6, 921-941, 2000.
and maybe (off the top of my head) also in
Seiler, F.A., and J.L. Alvarez, "Confounding Factors and the Use
of Epidemiological Data in Risk Assessment," Human and Ecological
Risk Assessment., 8, 607-610, 2002.
As Joe and I have neither the time nor the means, or the data for
such an effort, all we can do is point this fact out (time and
again!). It is a Model II linear regression that you need to do,
where both a stochastic incidence and a stochastic dose enter in
the fit.
Sometimes I think Joe and I are just shouting into the wind (like
the reporters covering the arrival of Isabel yesterday).
Cheers,
Fritz
*****************************************************
Fritz A. Seiler, Ph.D.
Sigma Five Consulting: Private:
P.O. Box 1709 P.O. Box 437
Los Lunas, NM 87031 Tome', NM 87060
Tel.: 505-866-5193 Tel. 505-866-6976
Fax: 505-866-5197
*****************************************************
*****************************************************
"This is the hour when democracy must justify
itself by capacity for effective decision, or risk
destruction or desintegration. Europe is dotted
with the ruins of right decisions taken too late."
"America's Responsibility in the Current Crisis"
Manifesto of the Christian Realists. May, 1940.
*******************************************************
II regression that is needed
-----Original Message-----
From: Frank Isackson [mailto:fisackson@earthlink.net]
Sent: Friday, September 19, 2003 12:38 PM
To: Mailing List for Risk Professionals
Subject: [riskanal] RE: If you do Science, use the Scientific Method!
On Friday, September 19, 2003, at 10:35 AM, Fritz A. Seiler wrote:
> If you fit a model to some data, let's assume a straight line model,
> you do a simple
> linear regressionand obtain a slope and a y-axis intercept and their
> errors. A look into
> any book on statistics will show you that you get two kinds of errors
> for the slope: One
> isthe errorfor the average slope and the other is the standard
> deviation of the average
> slope. The firstis theerror of the average risk for the population
> (slope = risk coefficient,
> in thelinear model), and the second is simply the distribution
> estimated for an assumed
> (n+1)st datapoint (that would yet have to be determined, but
While not intending for the discussion to descend into arguments about
statistics and causality let me amend Mr. Seiler's remarks by noting
that frequently in regression analyses the problem also involves an
estimate of the error of the independent variable, sometimes referred
to as the "Errors in Variables" problem or orthogonal regression.
Frank Isackson
---
The RISKANAL mailing list is intended to facilitate information exchange.
It is an unmoderated list. The sender of a message is responsible for
message content; there is no prior review or approval by the list
management.
You are currently subscribed to riskanal as: FASEILER@NMIA.COM
To unsubscribe send a blank email to leave-riskanal-2698D@lyris.pnl.gov
If you are experiencing problems with the
riskanal, please send email to jim.dukelow@pnl.gov
************************************************************************
You are currently subscribed to the Radsafe mailing list. To unsubscribe,
send an e-mail to Majordomo@list.vanderbilt.edu Put the text "unsubscribe
radsafe" (no quote marks) in the body of the e-mail, with no subject line.
You can view the Radsafe archives at http://www.vanderbilt.edu/radsafe/