fit_loess: Fit A Local Polynomial Regression Function
In bdpopt: Optimisation of Bayesian Decision Problems

Description Usage Arguments Details Value Author(s) See Also

Fit a local polynomial regression function to the estimated expected utility values obtained through simulation via JAGS by calling eval.on.grid. This is a generic function for S3 objects.

1	fit.loess(model, span = 0.75, degree = 2)

`model`	A model object obtained as the return value from `eval.on.grid`.
`span`	A parameter which controls the degree of smoothing.
`degree`	The degree of the polynomials to be used, normally 1 or 2.

This function calls loess in package stats to perform a regression. Note that the number of decision variables must be between 1 and 4, since this is the range supported by loess.

The formula passed as formula to loess has the form "y ~ x1 + x2" (for two decision variables, and correspondingly for any other number between 1 and 4). The span and degree arguments are passed on to loess as given. Further, surface = "direct" is used as a loess control value in order to allow for extrapolation for the fitted function. For the remaining arguments of loess, the default values are used.

A new simulation model object constructed from the object given as the first argument and the local polynomial regression results. The updated components in the new object are model$regression.fun and model$gpr.hyper.params (set to NA). See sim.model for a description of these components.

Sebastian Jobj<f6>rnsson jobjorns@chalmers.se

fit.gpr

bdpopt documentation built on May 2, 2019, 9:18 a.m.