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
GregMcMahon
Level III

Time series forecasting with two correlated variables

I'm sure there is a way to do this either in Fit Model, Time Series Analysis, or Time Series Forecast. I have two time series grouped by age bracket. Columns are population and health care expenditure. I can do some forecast modeling no problem with the data individually. But, I have a population boom in the 40-60 year age bracket, so when they become the 70-80 year age bracket, they will cause an increase in the burden of health care which won't be captured if I just do the forecast on the expenditure column itself.

 

Thanks for any advice and happy new year to everyone out there!

 

Best,

Greg

12 REPLIES 12
peng_liu
Staff

Re: Time series forecasting with two correlated variables

Please contact support@jmp.com , @GregMcMahon 

I don't recall whether this was an issue. I am not able to reproduce it in JMP15 or JMP14. Looks like the display is out of synch, the number editor boxes are lined up with their labels. They should have moved up by one row.

I can think of one reason: the column names. Recent versions of JMP use colon, quote, name, end quote, "n" to denote columns with unusual characters like dash.

peng_liu_0-1673014664643.png

Older versions of JMP use a Name function call

peng_liu_1-1673014793140.png

I am not sure how you modified the script, but I imagine that you have to modify both script and column names in the data table to make it run on older versions. I attach an updated data table for older version. If this version works, then I am pretty sure the trouble was due to column name. But if you have any doubt, please contact the tech support. I appreciate it.

In addition, after you get a newer version, you want to check out the Box-Cox transformation feature added in a recent release.

See Launch Time Series 

peng_liu_3-1673015236223.png

I am able to fit a plausible seasonal ARIMA (1,0,1)(1,1,0)19 model with lambda = 0.2174

peng_liu_4-1673015298698.png

 

 

GregMcMahon
Level III

Re: Time series forecasting with two correlated variables

Many thanks again Peng. It did not help so I wrote a lengthy message with attachments to support. I will keep you updated as I hear more. 

Best regards,

Greg

Re: Time series forecasting with two correlated variables

Update for you @GregMcMahon - the issue you reported was sent to JMP Development and a fix was made in the next major release of JMP ( JMP 18 ). We appreciate you taking the time to report the issue in detail; it helped to identify and find the correct solution.