Choose Language Hide Translation Bar
Highlighted
DL723
Level II

Rearranging an Fit model equation in an automated script

Hi,

   I have a simple problem but for some reason i'm drawing a blank on how to solve it haha.  I basically created a fit model for some data and had it output the formula into a column.  Now I want to set Y and all the factors except one to values and solve for the unknown.  In the example below, I want to solve for :RH assuming I have column data for all the other Variables.  I can do algebra by hand and probably use the desirability manuallly..but neither seem ideal for automating the process.  Any suggestions? Thanks in advance.

 

Example

 

(-10) + Match( :RD, 1, -0.2, 6, 0.2, . ) + Match( :O,
	4, -0.2,
	5, -2,
	6, 0.5,
	7, 0.1,
	10, -0.7,
	11, 0.6,
	12, -0.7,
	.
) + Match( :L, 1, -0.6, 2, 3, . ) + Match( :R,
	"B", 0.1,
	"T", -0.1,
	.
) + 6 * :RH + (:RH - 15) * Match( :RD, 1, -0.5, 6, 0.5, . )
2 REPLIES 2
Highlighted
phil_kay
Staff

Re: Rearranging an Fit model equation in an automated script

I think it would be easier to answer this via the prediction profiler. You can set the known factors to their values and find the setting of RH that gives the desired value of Y.  

Highlighted
jerry_cooper
Staff (Retired)

Re: Rearranging an Fit model equation in an automated script

You might also take a look at the Inverse Prediction function. Click the top most red triangle (assuming LS fit) after fitting your model and select Estimates->Inverse Prediction...  (Note that this option may be located under a different red triangle menu depending upon which fit personality you choose). This function is also accessible via scripting if you wish to automate it further.

Article Labels

    There are no labels assigned to this post.