This function acts as a switchboard for various regression modelling strategy,
and links these strategies to `compare`

.

1 | ```
strategy(g, method, int, int.adj, model)
``` |

`g` |
a data matrix to which a modelling strategy will be applied. |

`method` |
a list containing the a regression modelling strategy name and strategy-specific parameter values. |

`int` |
logical. If int is TRUE an intercept will be included in the regression model. |

`int.adj` |
logical. If int.adj is TRUE the intercept will be re-estimated after shrinkage is applied to the regression coefficients. |

`model` |
the type of regression model. Either "linear" or "logistic". |

For further details, see `compare`

.

The object returned by `strategy`

depends on the strategy selected in the
call to `compare`

. For example, if `method[1] = "lsq"`

, `strategy`

will return
a matrix of regression coefficients, whereas if `method[1] = "boot"`

, strategy
will return a list containing values returned by `bootval`

.

This function is not designed to be called directly, but acts as a workhorse
function for `compare`

Questions? Problems? Suggestions? Tweet to @rdrrHQ or email at ian@mutexlabs.com.

Please suggest features or report bugs with the GitHub issue tracker.

All documentation is copyright its authors; we didn't write any of that.