[Date Index] [Thread Index] [Author Index]

Re: Curve-fitting/data analysis question...

• To: mathgroup at smc.vnet.net
• Subject: [mg73355] Re: Curve-fitting/data analysis question...
• From: "sherifffruitfly" <sherifffruitfly at gmail.com>
• Date: Tue, 13 Feb 2007 07:03:31 -0500 (EST)
• References: <200702110615.BAA11578@smc.vnet.net><eqpecg$eme$1@smc.vnet.net>

On Feb 12, 2:09 am, "Chris Chiasson" <c... at chiasson.name> wrote:
> make the cutoff point a parameter in your regression?
>

So I ended up doing the following:

For each point p in my interval, I constructed two models - a linear
one using the points to the LEFT of p, and an exponential one using
the points to the RIGHT of p. I ran chi-squared to see which model p
belongs to. At the beginning, LEFT wins pretty much every time. When
RIGHT started winning is where I called the cut-off.

Does that seem to be a mathematically defensible approach?


Thanks for all the suggestions - the method above is something of an
amalgamation of all of them.

cdj

• References:
  • Curve-fitting/data analysis question...
    • From: "sherifffruitfly" <sherifffruitfly@gmail.com>