buildLM: Linear Model Interface
In bartzbeielstein/SPOT: Sequential Parameter Optimization Toolbox

Description Usage Arguments Value Examples

This is a simple wrapper for the lm function, which fits linear models. The purpose of this function is to provide an interface as required by SPOT, to enable modeling and model-based optimization with linear models. The linear model is build with main effects. Optionally, the model is also subject to the AIC-based stepwise algorithm, using the step function from the stats package.

1	buildLM(x, y, control = list())

`x`	matrix of input parameters. Rows for each point, columns for each parameter.
`y`	one column matrix of observations to be modeled.
`control`	list of control parameters, currently only with parameters `useStep` and `formula`. The `useStep` boolean specifies whether the `step` function is used. The `formula` is passed to the lm function. Without a formula, a second order model will be built.

an object of class "spotLinearModel", with a predict method and a print method.

## Test-function:
braninFunction <- function (x) {	
	(x[2]  - 5.1/(4 * pi^2) * (x[1] ^2) + 5/pi * x[1]  - 6)^2 + 
10 * (1 - 1/(8 * pi)) * cos(x[1] ) + 10
}
## Create design points
set.seed(1)
x <- cbind(runif(20)*15-5,runif(20)*15)
## Compute observations at design points (for Branin function)
y <- as.matrix(apply(x,1,braninFunction))
## Create model
fit <- buildLM(x,y,control = list(algTheta=optimLHD))
## Print model parameters
print(fit)
## Predict at new location
predict(fit,cbind(1,2))
## True value at location
braninFunction(c(1,2))