Performs a stepwise forward variable/model selection using the Wilk's Lambda criterion.

```
greedy.wilks(X, ...)
# S3 method for default
greedy.wilks(X, grouping, niveau = 0.2, ...)
# S3 method for formula
greedy.wilks(formula, data = NULL, ...)
```

X

matrix or data frame (rows=cases, columns=variables)

grouping

class indicator vector

formula

formula of the form ‘`groups ~ x1 + x2 + ...`

’

data

data frame (or matrix) containing the explanatory variables

niveau

level for the approximate F-test decision

...

further arguments to be passed to the default method, e.g. `niveau`

A list of two components, a `formula`

of the form ‘`response ~ list + of + selected + variables`

’,
and a data.frame `results`

containing the following variables:

the names of the variables in the final model in the order of selection.

the appropriate Wilks' lambda for the selected variables.

the approximated F-statistic for the so far selected model.

the appropriate p-value of the F-statistic.

the approximated F-statistic of the partial Wilks's lambda (for comparing the model including the new variable with the model not including it).

the appropriate p-value of the F-statistic of the partial Wilk's lambda.

A stepwise forward variable selection is performed. The initial model is defined by starting with the variable which separates the groups most. The model is then extended by including further variables depending on the Wilk's lambda criterion: Select the one which minimizes the Wilk's lambda of the model including the variable if its p-value still shows statistical significance.

Mardia, K. V. , Kent, J. T. and Bibby, J. M. (1979), *Multivariate analysis*,
Academic Press (New York; London)

# NOT RUN { data(B3) gw_obj <- greedy.wilks(PHASEN ~ ., data = B3, niveau = 0.1) gw_obj ## now you can say stuff like ## lda(gw_obj$formula, data = B3) # }