diff --git a/sklearn/metrics/__init__.py b/sklearn/metrics/__init__.py
index 93d21a146619a..369b08270d35d 100644
--- a/sklearn/metrics/__init__.py
+++ b/sklearn/metrics/__init__.py
@@ -14,6 +14,7 @@
 from .ranking import roc_curve
 from .ranking import dcg_score
 from .ranking import ndcg_score
+from .ranking import gini
 
 from .classification import accuracy_score
 from .classification import classification_report
diff --git a/sklearn/metrics/ranking.py b/sklearn/metrics/ranking.py
index 9755732a4f910..34194c10d9d40 100644
--- a/sklearn/metrics/ranking.py
+++ b/sklearn/metrics/ranking.py
@@ -858,3 +858,33 @@ def ndcg_score(y_true, y_score, k=5):
         scores.append(actual / best)
 
     return np.mean(scores)
+
+
+def gini(y_true, y_score):
+    """ Compute Gini coefficient 
+
+    Compute the Gini coefficient as Gini = 2 × AUC - 1 [1].
+
+    Parameters
+    ----------
+    
+    y_true : array, shape = [n]
+            Actual target values for X.
+
+    y_score : array, shape = [n]
+        Probability estimates of the positive class.
+
+    Returns
+    -------
+    gini : float
+
+    References
+    ----------
+    .. [1] David J. Hand  and Robert J. Till (2001).
+            A Simple Generalisation of the Area Under the ROC Curve for
+            Multiple Class Classification Problems. In Machine Learning, 45,
+            pp.171–186 (Kluwer Academic Publishers).
+
+    """
+
+    return 2*roc_auc_score(y_true, y_score)-1
diff --git a/sklearn/metrics/scorer.py b/sklearn/metrics/scorer.py
index b1f01c1a18e1b..4a37dbee7b07c 100644
--- a/sklearn/metrics/scorer.py
+++ b/sklearn/metrics/scorer.py
@@ -514,7 +514,7 @@ def make_scorer(score_func, greater_is_better=True, needs_proba=False,
 log_loss_scorer = make_scorer(log_loss, greater_is_better=False,
                               needs_proba=True)
 log_loss_scorer._deprecation_msg = deprecation_msg
-
+gini_scorer = make_scorer(gini, greater_is_better=True)
 
 # Clustering scores
 adjusted_rand_scorer = make_scorer(adjusted_rand_score)