Programmed a solution in Statistica Basic that transforms the raw data
into the proper form for the "Logit" tool: codes and [frequency]
counts. Then, used the Logit tool to fit a curve to the data
using "maximum likelyhood" prediction method. Looked at the 2D
function plot and it looked reasonable. However, when I decrease the
ration of ones to zeros (thus decreasing the probability) at the
beginning of the data set (time ordered), the function plot is way off
in the center where there are entirely all ones. Is there a better
model to use? One that will not go "haywire" when data is changed at
the beginning or end of the data set?
In article <88udoc$559$[EMAIL PROTECTED]>,
[EMAIL PROTECTED] wrote:
> In article <[EMAIL PROTECTED]>,
> [EMAIL PROTECTED] wrote:
> > On Tue, 15 Feb 2000 16:34:54 GMT, [EMAIL PROTECTED] wrote:
> >
> > > I have a data set with two variables (with some correlation
between
> > > them,
> > > such that as y increases the probability of x decreases). First,
> I'm
> > > not a
> > > statician, so bare with me on this. One variable, say "x", is a
> > > boolean and
> > > represents that an event has occurred. The question is "at what y
> value
> > > does the probablility of event x become <such-and-such>"?
> Furthermore,
> > > is
> > > it necessary to create "bins" (ie., divide y into bins, so that
you
> have
> >
> > A logistic regression gives you a fitted curve, based on a model
which
> > is fits 'growth' scenarios (among others) neatly, and is not very
far
> > off from a cumulative normal distribution. Have you tried that?
> >
> > --
> > Rich Ulrich, [EMAIL PROTECTED]
> > http://www.pitt.edu/~wpilib/index.html
> >
> I've looked at it in Statistica (Logit tool), but there seems to be a
> number of variables that are being asked for. Would you have any
> advice on using Logit? Also, I'm just now reading up on discrete
> random variables and the various distributions associated with them.
> It mentions the cumulative probability function (is this the same
> as/related to the cumulative normal distribution?). I am interested
in
> two things: (1) increase knowledge in statistics, and (2) how to use
> Statistica. I'll probably let you know by tomorrow 2/23 how it's
going.
>
> Thanks,
>
> grows
>
> Sent via Deja.com http://www.deja.com/
> Before you buy.
>
Sent via Deja.com http://www.deja.com/
Before you buy.
===========================================================================
This list is open to everyone. Occasionally, less thoughtful
people send inappropriate messages. Please DO NOT COMPLAIN TO
THE POSTMASTER about these messages because the postmaster has no
way of controlling them, and excessive complaints will result in
termination of the list.
For information about this list, including information about the
problem of inappropriate messages and information about how to
unsubscribe, please see the web page at
http://jse.stat.ncsu.edu/
===========================================================================
