diff --git a/pandas/core/series.py b/pandas/core/series.py
index 7400aa5bde2e7..ac8fada304cf9 100644
--- a/pandas/core/series.py
+++ b/pandas/core/series.py
@@ -1513,6 +1513,42 @@ def autocorr(self):
         """
         return self.corr(self.shift(1))
 
+    def autocov(self, j=1, unbiased=False):
+        """
+        j-lag autocovariance
+
+        Parameters
+        ----------
+        j: int, default 1
+            Periods to lag the covariance calculation by
+        unbiased : boolean, default False
+            If true return an unbiased estimator of the autocovariance
+
+        See Also
+        --------
+        statsmodels.tsa.statstools.acovf for autocovariance function, which
+        returns an array of lagged autocovariances
+
+        Returns
+        -------
+        autocov : float
+        """
+        n = len(self)
+
+        if abs(j) >= n or n == 0:
+            return np.nan
+
+        this = self - self.mean()
+        shifted = this.shift(-j)
+
+        if unbiased:
+            d = n - j
+        else:
+            d = n
+
+        return float(np.correlate(this[:n - j].values,
+                                  shifted[:n - j].values)) / d
+
     def clip(self, lower=None, upper=None, out=None):
         """
         Trim values at input threshold(s)
diff --git a/pandas/tests/test_series.py b/pandas/tests/test_series.py
index 9474f6061f936..1826916a985ba 100644
--- a/pandas/tests/test_series.py
+++ b/pandas/tests/test_series.py
@@ -1841,6 +1841,19 @@ def test_cov(self):
         cp[:] = np.nan
         self.assert_(isnull(cp.cov(cp)))
 
+    def test_autocov(self):
+
+        ts = Series([1, 2, 3] * 2)
+
+        #too big of lag
+        self.assert_(np.isnan(ts.autocov(j=len(ts) + 1)))
+
+        #test calculations
+        self.assertAlmostEqual(ts.autocov(j=0), 2.0/3)
+        self.assertAlmostEqual(ts.autocov(j=1), -1.0/6)
+        self.assertAlmostEqual(ts.autocov(j=2), -1.0/3)
+
+
     def test_copy(self):
         ts = self.ts.copy()