mda: Mixture Discriminant Analysis.

In Fuzzy-Logix/AdapteR: Converts R syntax to SQL to provide fast analytics using in-database C++ library DB Lytix.

Description

mda performs Mixture Discriminant Analysis on FLTable objects. The DB Lytix function called is FLMDA. Performs Mixture Discriminant Analysis and stores the results in predefined tables.

Usage

mda(formula, data = list(), ...)

Arguments

formula	A symbolic description of model to be fitted
data	An object of class FLTable or FLTableMD.
subclasses	Number of subclasses.
iter	Maximum number of iterations for expectation maximization.
init	Initialization method for each obs' latent variable Prob(x is in the subclass of class) 1 = Assign weight of 1 to a random subclass of its class; 0 otherwise.
hypothesis	Number of hypotheses to run simultaneously

Details

limitations: no dollar operator access to means, precent.explained, values, plot((lack of data of discriminant space).

Value

mda returns an object of class FLLDA

See Also

mda for R reference implementation.

Examples

deeptbl <- FLTable("tblMDA","ObsID", "VarID", "Num_Val")
flmod <- mda(formula = a~.,data = deeptbl)
vpred <- predict(flmod); flmod$N
FLMDA performs mixed discriminant analysis. For the training data, MDA divides each
class into a number of artificial subclasses. It calibrates the mixture of Gaussians
and the mixing probability by maximizing the log-likelihood with expectation maximization.

Fuzzy-Logix/AdapteR documentation built on May 6, 2019, 5:07 p.m.