Auto-suggest helps you quickly narrow down your search results by suggesting possible matches as you type.
Showing results for
Search instead for
Did you mean:
Our World Statistics Day conversations have been a great reminder of how much statistics can inform our lives. Do you have an example of how statistics has made a difference in your life? Share your story with the Community!
I need some help! It was very early this morning when I tried to analyze some data with JMP Functional Data Explorer. I used the data with functions as columns and I analyzed them with B-splines. I got very nice results, I could see directly the sigmoidal curves on the plots and also the PCA was fine. I tried again later but it did not work. I updated from JMP 15.0.0 to the new version but it still did not produce what I expected. I tried P-splines but nothing. I am attaching the jmp data table with a script. It is like the fitting procedure stops at some x value so that just a subset of the data is well fitted. The PCA results are either strange (both components identical) or ‘not available’. Strangely when I look at the diagnostic plots everything looks perfect. If I was awake in the early morning and I saw JMP producing beautiful results my guess is that JMP uses some starting points for numerical optimization that it reminds from previous runs… Is there a way to change this initial condition? It may be useful to log transform the x scale to better visualize the data. I tried also to change settings in preferences without success.
Marcello, It looks like FDE is not handling the Log setting on the X axis well. So instead, first try to create a transformed column Log[X] (right click X, New Formula Column, Transform, Log). Then, specify Log[X] as X and model the data. Hope this helps!
Hello Ryan, Thank you very much for your response. I actually tried this solution and I was able to overcome the problem. The fitting worked perfectly. Based on this experience, I think it would be very useful to include a transformation button in FDE that gives the possibility to transform not only the y but also the x (as for the fit special platform).