Date:         Thu, 17 Jul 2003 13:47:23 GMT
Reply-To:     julierog@ix.netcom.com
Sender:       "SAS(r) Discussion" <SAS-L@LISTSERV.UGA.EDU>
From:         Roger Lustig <trovato@VERIZON.NET>
Subject:      Re: Use SAS to evaluate equation
Content-Type: text/plain; charset=us-ascii; format=flowed

Rick:

If you've had two or three weeks of calculus, you can do that with
pencil and paper--or perhaps a few punches of your calculator.

At the maximum, the slope of the curve is zero.  The slope of the curve
is the derivative--which, in this case, is a quadratic equation:

b + 2cx + 3dx**2.

Set that to zero and solve, using the quadratic formula.  It should give
you the two values of X for which the derivative is zero, i.e., a local
max or min exists.  The sign of parameter D will tell you which one is
which.

Best,
Roger Lustig

Rick wrote:
> Howdy -
>
> I've just run a PROC REG, and have the constants for a cubic equation.
>
> Example:   a + bx + cx**2 +dx**3
>
> I have a dataset called REG where a,b,c,and d are defined. Now I want
> to evaluate the equation to find the MAX point in the curve.
>
> The obvious way to do this is to create a new dataset, calculate a
> number of points, then do a PROC MEANS to find the max.... TONS of
> work, especially if I want accuracy...
>
> Is there some easier way to do this???
>
> On my calculator I can do something like this:
> MAX(a + bx +cx**2 + dx**3), -.5, .5
> parm[equation],[min range,  max range]
>
> Thanks,
>
> -Rick