| Logistic model in COMFIT function [message #93810] |
Tue, 25 October 2016 13:37  |
bedroom330
Messages: 1
Registered: October 2016
|
Junior Member |
|
|
Hi,
I am currently using the logistic model in COMFIT. The results are good; however, when I read the equation IDL uses for logistic model, I got confused. The equation is: y=1/(a0a1^x+a2). I can not understand why a0 and a1 are necessary? They can be combined as one empirical parameter. How does each of the empirical parameter work in this IDL function?
If you can provide any thought, I will greatly appreciate.
Thank you.
Sincerely,
Nick
|
|
|
|
| Re: Logistic model in COMFIT function [message #93813 is a reply to message #93810] |
Wed, 26 October 2016 07:24  |
Dick Jackson
Messages: 347
Registered: August 1998
|
Senior Member |
|
|
On Tuesday, 25 October 2016 13:37:22 UTC-7, bedro...@gmail.com wrote:
> Hi,
>
> I am currently using the logistic model in COMFIT. The results are good; however, when I read the equation IDL uses for logistic model, I got confused. The equation is: y=1/(a0a1^x+a2). I can not understand why a0 and a1 are necessary? They can be combined as one empirical parameter. How does each of the empirical parameter work in this IDL function?
>
> If you can provide any thought, I will greatly appreciate.
>
> Thank you.
>
> Sincerely,
> Nick
Hi Nick,
To make that equation clearer, to show the order of operations when it is evaluated (exponents before multiplication), it could be written as:
y = 1 / ( ( a0 * (a1^x) ) + a2 )
Does that help to show how a0 and a1 are quite distinct?
Cheers,
-Dick
Dick Jackson Software Consulting Inc.
Victoria, BC, Canada --- http://www.d-jackson.com
|
|
|
|
| Re: Logistic model in COMFIT function [message #93814 is a reply to message #93810] |
Wed, 26 October 2016 07:28 |
dcleuclidwl
Messages: 6
Registered: July 2016
|
Junior Member |
|
|
Perhaps your confusion comes from not taking into account that exponents take precedence over multiplication?
So y=1/(a0a1^x+a2) is the same as y = 1/(a0*(a1^x) + a2)
On Tuesday, October 25, 2016 at 4:37:22 PM UTC-4, bedro...@gmail.com wrote:
> Hi,
>
> I am currently using the logistic model in COMFIT. The results are good; however, when I read the equation IDL uses for logistic model, I got confused. The equation is: y=1/(a0a1^x+a2). I can not understand why a0 and a1 are necessary? They can be combined as one empirical parameter. How does each of the empirical parameter work in this IDL function?
>
> If you can provide any thought, I will greatly appreciate.
>
> Thank you.
>
> Sincerely,
> Nick
|
|
|
|
comp.lang.idl-pvwave archive
Members
Search
Help
Login
Home
