|
|
|
|
|
|
|
|
CTK Exchange
|
alexb
Charter Member
2391 posts |
Jun-05-09, 04:11 PM (EST) |
|
3. "RE: Right Tailed Distribution Function"
In response to message #2
|
You may start with something like fc(x) = (log(x+c) - log(c)) / (x + c) This is a positive function (for x > 0) that aproaches 0 at infinity, is 0 at 0, and has a single maximum. This is true for any c > 0. Now you can consider, say, n·fc(m·x) with three parameters to choose the most suitable function. For b, you must be able to choose one of the three or some combination thereof by relating their values. Actually, the base of the logarithm is a fourth parameter. |
|
Alert | IP |
Printer-friendly page |
Reply |
Reply With Quote | Top |
|
|
|
Marcus Bizony
guest
|
Jun-21-09, 10:43 AM (EST) |
|
4. "RE: Right Tailed Distribution Function"
In response to message #0
|
Since you ask for a smooth curve perhaps you will not like to use a modulus. Otherwise the Cupid's Bow function Mod(2bx/(x^2+1))seems to satisfy your requirements |
|
Alert | IP |
Printer-friendly page |
Reply |
Reply With Quote | Top |
|
|
You may be curious to have a look at the old CTK Exchange archive. Please do not post there.
Copyright © 1996-2018 Alexander Bogomolny
|
|