diff --git a/NAMESPACE b/NAMESPACE
index 4bc04ca..6773b62 100644
--- a/NAMESPACE
+++ b/NAMESPACE
@@ -2,6 +2,7 @@
 
 export(mnre_dim_and_class_to_index)
 export(mnre_expand_matrix)
+export(mnre_fit)
 export(mnre_fit_sparse)
 export(mnre_make_covar)
 export(mnre_simulate_ev_data)
diff --git a/R/main.R b/R/main.R
index 3fdc4a6..4c22a8c 100644
--- a/R/main.R
+++ b/R/main.R
@@ -208,28 +208,41 @@ nd_min_fun <- function(ev) {
                            data=ev$fr, family='binomial')
     fe <- fixed_effects <- (glf$X)
     re <- random_effects <- Matrix::t(glf$reTrms$Zt)
-    y <- matrix(ev$y, ncol=1)
+    
+    y <- matrix(ev$fr[,all.vars(ev$frm)[[1]]], ncol=1)
     k_class <- max(y)
     k <- max(y)
-    Lind = ev$Lind
+    Lind = glf$reTrms$Lind
     
     s = 'mval '
     for (v in mval) {
       s = sprintf("%s %.4e ", s, v)
     }
-    message(s)
+
+    if (ev$verbose > 0) {
+      message(s)      
+    }
+
     
     theta_mat <- matrix(mval, ncol=k_class)
     
     fe2 <- Matrix::Matrix(fe, sparse = TRUE)
     
+    if (! "beta_re" %in% names(ev)) {
+      ev$beta_re <- matrix(rnorm(ncol(re) * k_class, 0, 0.2), ncol=k_class)
+    }
+
+    if (! "beta_fe" %in% names(ev)) {
+      ev$beta_fe <- matrix(rnorm(ncol(fe) * k_class, 0, 0.2), ncol=k_class)      
+    }    
+    
     beta_re <- ev$beta_re
     beta_fe <- ev$beta_fe
     
-    # beta_re <- matrix(rep(0, ncol(re) * k_class), ncol=k_class)
-    # beta_fe <- matrix(rep(0, ncol(fe) * k_class), ncol=k_class)
-    #
-    message("starting beta ", beta_fe[[1]], " ", beta_re[[1]])
+    if (ev$verbose > 0) {
+      message("starting beta ", beta_fe[[1]], " ", beta_re[[1]])
+    }
+    
     zz <- mnre_fit_sparse(fe2, re, y, theta_mat, Lind, beta_fe, beta_re, verbose = ev$verbose)
     
     ev$beta_fe <<- zz$beta_fixed
diff --git a/R/mnre_fit.R b/R/mnre_fit.R
new file mode 100644
index 0000000..875f88d
--- /dev/null
+++ b/R/mnre_fit.R
@@ -0,0 +1,16 @@
+
+#' @export 
+mnre_fit <- function(frm, data, verbose=0, off_diagonal=0.0) {
+  ev <- list() 
+  ev$frm <- frm
+  ev$fr <- data
+  ev$verbose <- verbose
+  
+  nlev <- length(all.vars(ev$frm))
+  mval <- rep(1,  (nlev-1) * max(ev$fr$y))   
+
+  nf <- nd_min_fun(ev)
+  
+  ans = optim(mval, nf, method = "L-BFGS", lower=1e-8)
+
+}
\ No newline at end of file