Re: hyperbolic distribution
- From: "[Mr.] Lynn Kurtz" <kurtzDELETE-THIS@xxxxxxx>
- Date: Sun, 28 May 2006 23:20:09 GMT
On 28 May 2006 15:25:36 -0700, "Martin" <mgcqso@xxxxxxxxxxx> wrote:
Hey folks, remember me? I had that problem with the exponential
function. In continuing to work on that, I have realized that I
actually need an asymmetric spike, one with the left side pulled up and
out. I surfed the net and found something called the "hyperbolic
distribution," a variation on the exponential function. It is at
http://www.2dcurves.com/exponential/exponentialhd.html
The function I suggested was more complicated than the others but it
has more flexibility also. It was of the form:
y = k { 1 / 1(1+exp(-m(x-a))) - 1 / (1+exp(-n(x-b))) }
You can choose a and b near your peak, k to set your peak height, and
adjust m and n to vary the shape on each side.
--Lynn
.
- Follow-Ups:
- Re: hyperbolic distribution
- From: Martin
- Re: hyperbolic distribution
- References:
- hyperbolic distribution
- From: Martin
- hyperbolic distribution
- Prev by Date: Re: naive question from a non-mathematician
- Next by Date: Re: looking for good undergrad calculus text
- Previous by thread: hyperbolic distribution
- Next by thread: Re: hyperbolic distribution
- Index(es):
Relevant Pages
|