[R] Package for error analysis

[S Ellison]

> I'm doing error analysis of predictive models and I need to calculate global
> error, this is, I need to calculate the resultant error from propagation of
> indirect measurements errors.

If the problem is propagation of variance on inputs through a known function, you could look at uncert() in the metRology package.
That basically does first order error propagation using analytic (if available) or numerical differentiation of the function, or (via uncertMC()) Monte Carlo simulation to get error distributions (which I would recommend for larger variance over first order calculations).

I wouldn't exactly call it statistics - it's based on recommendations for measurement science and they assume a pretty simple deterministic model and that has little or nothing to do with model fitting and inference.

S Ellison

> -----Original Message-----
> From: R-help [mailto:r-help-bounces at r-project.org] On Behalf Of Catarina
> Silva
> Sent: 04 February 2016 09:57
> To: R mailling list
> Subject: [R] Package for error analysis
> 
> Hi,
> 
> I'm doing error analysis of predictive models and I need to calculate global
> error, this is, I need to calculate the resultant error from propagation of
> indirect measurements errors.
> 
> I know a little about propagation error theory, using derivation formulas to
> calculate it, but I want to know if exists some package to do it.
> 
> 
> 
> Ty
> 
> Catarina
> 
> 
> 
> 
> 	[[alternative HTML version deleted]]
> 
> ______________________________________________
> R-help at r-project.org mailing list -- To UNSUBSCRIBE and more, see
> https://stat.ethz.ch/mailman/listinfo/r-help
> PLEASE do read the posting guide http://www.R-project.org/posting-guide.html
> and provide commented, minimal, self-contained, reproducible code.


*******************************************************************
This email and any attachments are confidential. Any use...{{dropped:8}}