Formula for a Fractional Factorial Design

DESCRIPTION:

Returns a model formula that can be used to fit (in general) an effect saturated model using the function fac.aov. This function is used to generate the default formula in a typical call to fac.aov.

USAGE:

formula.fac.design(object, response) 

REQUIRED ARGUMENTS:

object
an object of class "fac.design", typically created using design.digest, fac.design, or as.fac.design.

OPTIONAL ARGUMENTS:

response
the response variable. The default is to use the first non-factor in the fac.design object.

VALUE:

an object of class formula that typically will specify an effect saturated model for the fac.design object.

DETAILS:

The effect saturated model is generated as a fully crossed model if there are fewer than seven factors (e.g., ` y~A * B * C * D') and a crossed model of order 3 for seven or more factors (e.g., ` y~.^3' ). The presence of a factor named "Quad" indicates that there are center points and a term "Quad" that does not interact with any model terms will be added to the model. If there is a blocking factor, it will be added as the first term in the formula (e.g., `y ~ Block1 + A*B*C*D + Quad').

SEE ALSO:

, , , .

EXAMPLES:

a.design <- design.digest('ff0816') 
formula(a.design) 
formula(buffer.df) 
# the following two statements are equivalent 
fac.aov(formula(buffer.df),data=buffer.df)  
fac.aov(buffer.df)