topic Re: Create a RSM using already collected historical data in Discussions

Create a RSM using already collected historical data

soumyar — Mon, 05 Jun 2017 06:23:20 GMT

Hi,

I have a dump of historical values for the explanatory and response variables. How do i build an RSM given this data? Can i build a custom design based on data already collected for a series of runs?

Re: Create a RSM using already collected historical data

txnelson — Mon, 05 Jun 2017 06:31:49 GMT

Yes. You can build your custom design that matches your historical data, and then once the model is built, you just cut and paste the historical data into the design

Re: Create a RSM using already collected historical data

soumyar — Mon, 05 Jun 2017 06:41:37 GMT

Thanks for the reply! Any material that I can refer to for building the custom design?

Re: Create a RSM using already collected historical data

txnelson — Mon, 05 Jun 2017 06:44:02 GMT

Help==>Books==>Design of Experiments Guide

Help==>Tutorials==>DOE Tutorial

Re: Create a RSM using already collected historical data

soumyar — Mon, 05 Jun 2017 06:52:36 GMT

Creating custom factorial design can handle historical data as random as the below sample data?

-> X1 - X5 - Independent Variable

-> Y - Response Variable

Run Order	X1	X2	X3	X4	X5	Y
1	69	0	430	-3	11	23
2	73	0	421	-2	12	21
3	86	1	426	-1	12	27
4	83	0	428	-3	13	29
5	58	1	391	-3	11	24
6	86	0	385	-4	13	28
7	74	1	446	-4	14	29
8	85	0	405	-4	13	26
9	93	0	368	-2	15	21
10	75	1	382	-1	12	28
11	93	1	414	-3	11	27
12	98	1	422	-2	14	21
13	63	1	397	-2	15	29
14	73	1	355	-2	12	28
15	61	0	420	-4	11	23
16	92	2	390	-2	15	22
17	75	2	427	-2	14	25
18	93	0	379	-1	10	28
19	78	2	431	-2	10	30
20	77	0	390	-1	15	22

Re: Create a RSM using already collected historical data

txnelson — Mon, 05 Jun 2017 06:54:50 GMT

It should not be an issue

Re: Create a RSM using already collected historical data

Peter_Bartell — Mon, 05 Jun 2017 17:07:44 GMT

Maybe I'm missing something but since you already have the x and y matrix from your historical data, why do you need to create a custom design when your primary goal is for RSM model evaluation? You can just use the JMP data table for your historical data and then go straight to the Fit Model platform, and then pick the appropriate Fit Model personality, effect specification, etc. You can still use the Evaluate Design platform on your design matrix to evaluate for Power, correlation among effects, etc.

And how exactly are you going to create the custom design? I would find it highly unlikely that the custom design platform for an I optimal design is going to have a set of treatment combinations that you can find within your historical data collection of combinations?

I'd just be mindful of correlation among the predictor variables for personalities such as Standard Least Squares. One primary advantage of DOE is to AVOID this problem...but historical data doesn't usually come from a designed experiment...you get what you get...and multicollinarity/correlation among predictor variables is often present. All is not lost if you have substantial amounts of multicollinearity...there are still modeling personalities in JMP (like partial least squares) and JMP Pro (the penalized regression methods in the Generalized Regression personality) which are useful in this eventuality.