Add Bayesian Confidence Interval for Bivariate Fit Spline

[shaira]

JMP's Bivariate Fit Spline has nice capabilities and customization but lacks option for addition of confidence intervals. It would be good to have a Bayesian Confidence Interval visualization whose equation we can also export as column formulas. We can have option to set the confidence interval to 95%, 90%, etc.

[SamGardner]

@shaira thank you for the suggestion.  In Bivariate we currently do not provide any capability to show or calculate confidence intervals for smoothed flexible curve fits. 

Can you explain in more detail what methodology you think would be best to do this, and also why this is needed and important to you?  I found this reference for Bayesian Confidence intervals for smoothing splines 

• Wahba, G. (1983). “Bayesian 'Confidence Intervals' for the Cross Validated Smoothing Spline.” Journal of the Royal Statistical Society, Series B 45:133–150.

[SamGardner]

@shaira have you been able to consider this and can you provide more detail on the methodology you think would be best.

[SamGardner]

We have not received a response to the request for more information, so we are closing this request.  If you have additional information to share, we can re-open this and investigate further.  

[bob_stine]

The Graph Builder platform shows confidence intervals for the spline in the figure.  Why not share that technique for a confidence interval with the Fit Y by X platform?  You have the code, so add the same type of intervals in Fit Y by X when a user adds the spline fit.