# stepfit: step-wisd multivariate model fitting In MVB: Mutivariate Bernoulli log-linear model

## Description

stepwise fit multivariate log-linear Bernoulli model using Newton-Raphson algorithm.

## Usage

 ```1 2 3 4 5``` ```stepfit(x, y, maxOrder = 2, output = 0, direction = c("backward", "forward"), tune = c("AIC", "BIC", "GACV", "BGACV"), start = NULL) ```

## Arguments

 `x` input design matrix. `y` output binary matrix with number of columns equal to the number of outcomes per observation. `maxOrder` maximum order of interactions to be considered in outcomes. `output` with values 0 or 1, indicating whether the fitting process is muted or not. `direction` the mode of stepwise search and default is backward. `tune` tuning approach, available methods including AIC, BIC, GACV, BGACV. `start` starting object of type mvbfit.

## Details

The `stepfit` utilize the class structure of the underlying C++ code and stepwisd fitted the model with Newton-Raphson algorithm.

## Value

An object of class `mvbfit`, for which some methods are available.

`mvblps`, `unifit`, `stepfit`, `mvb.simu`
 ``` 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18``` ```# fit a simple MVB log-linear model n <- 1000 p <- 5 kk <- 2 tt <- NULL alter <- 1 for (i in 1:kk) { vec <- rep(0, p) vec[i] <- alter alter <- alter * (-1) tt <- cbind(tt, vec) } tt <- 1.5 * tt tt <- cbind(tt, c(rep(0, p - 1), 1)) x <- matrix(rnorm(n * p, 0, 4), n, p) res <- mvb.simu(tt, x, K = kk, rep(.5, 2)) fitMVB <- mvbfit(x, res\$response, output = 1) ```