cancel
Showing results for 
Show  only  | Search instead for 
Did you mean: 
Browse apps to extend the software in the new JMP Marketplace
Choose Language Hide Translation Bar
MikeKim
Level IV

Please suggest the Method

Hi, 

Please suggest the analysis method for the following data figure.

X1   X2   X3   X4   X5   Y

21   33    2     66    81   90

17   39   10    17    90   120

5     55   6      51    95   106

...

I am trying to figure out which X is related to the Y response.

if there is any X, what would the equation be?

 

Thank you for helping.

1 ACCEPTED SOLUTION

Accepted Solutions
drdrf
Level III

Re: Please suggest the Method

Hello @MikeKim .

 

Unfortunately I think the short answer to your question is that more data is needed to be able to determine which Xs are related to the Y.

 

Do you have a reason to believe that only one X will be related to the response, or might several Xs have and influence on Y?

 

A scatterplot of the data provides a few clues as to what might be going on, but I would avoid drawing specific conclusion based only on three runs.

 

drdrf_0-1710508262212.png

Firstly, X1 seems to correlate with X2. X3 also seems to correlate with X4. Is this fundamental to the way the system operates?  If these correlations are expected and/or can be confirmed by further data, this might allow a predictive model to be built with less data than would be required for five independent Xs. A model could be built with only one of each correlating pair, or a Principal Component approach could be used.

 

There is also an indication Y might correlate with X3 and X4.  This could be investigated if more data was available.

 

The best path forward would depend on what type of system is being studied.  Will only unstructured data be available or can designed experiments be conducted to answer the question?

View solution in original post

4 REPLIES 4
drdrf
Level III

Re: Please suggest the Method

Hello @MikeKim .

 

Unfortunately I think the short answer to your question is that more data is needed to be able to determine which Xs are related to the Y.

 

Do you have a reason to believe that only one X will be related to the response, or might several Xs have and influence on Y?

 

A scatterplot of the data provides a few clues as to what might be going on, but I would avoid drawing specific conclusion based only on three runs.

 

drdrf_0-1710508262212.png

Firstly, X1 seems to correlate with X2. X3 also seems to correlate with X4. Is this fundamental to the way the system operates?  If these correlations are expected and/or can be confirmed by further data, this might allow a predictive model to be built with less data than would be required for five independent Xs. A model could be built with only one of each correlating pair, or a Principal Component approach could be used.

 

There is also an indication Y might correlate with X3 and X4.  This could be investigated if more data was available.

 

The best path forward would depend on what type of system is being studied.  Will only unstructured data be available or can designed experiments be conducted to answer the question?

MikeKim
Level IV

Re: Please suggest the Method

I apologize for late.

Actually, the data is mock. 

What I am asking for was for that data type, what analysis would be applicable.

I had concluded that I start with the multivariate correlation as you showed me.

Thank you for your support and let me know if there is any other screening method.

My current strategy is, 

1. multivari-correlation

2. respective regression analysis (if corr is exist)

 

Re: Please suggest the Method

Two very useful screening methods are Response Screening and Predictor Screening. Response Screening is very useful for large data sets where you are looking for only linear relationships, and Predictor Screening is very useful for when you suspect interactions or curvature.

 

Jed_Campbell_0-1711117791591.png

 

MikeKim
Level IV

Re: Please suggest the Method

Appreciates for your answer.

But I have no idea how to further analyze the results of the Response screening.

Quite difficult to interpret the results and no background knowledge of this platform. 

Sorry this. 

But thank you!