diff --git a/requirements.txt b/requirements.txt
index aaf97d1b7..a347a9f33 100644
--- a/requirements.txt
+++ b/requirements.txt
@@ -19,7 +19,7 @@ gunicorn==19.8.0
 humanize==0.5.1
 idna==2.6
 markdown==2.6.11
-pandas==0.22.0
+pandas==0.23.1
 parsedatetime==2.0.0
 pathlib2==2.3.0
 polyline==1.3.2
diff --git a/setup.py b/setup.py
index 0e7243f35..bb216886a 100644
--- a/setup.py
+++ b/setup.py
@@ -76,7 +76,7 @@ setup(
         'humanize',
         'idna',
         'markdown',
-        'pandas',
+        'pandas>=0.18.0',
         'parsedatetime',
         'pathlib2',
         'polyline',
diff --git a/superset/viz.py b/superset/viz.py
index 6b3c6d843..ee423b4d8 100644
--- a/superset/viz.py
+++ b/superset/viz.py
@@ -1175,15 +1175,14 @@ class NVD3TimeSeriesViz(NVD3Viz):
 
         if rolling_type in ('mean', 'std', 'sum') and rolling_periods:
             kwargs = dict(
-                arg=df,
                 window=rolling_periods,
                 min_periods=min_periods)
             if rolling_type == 'mean':
-                df = pd.rolling_mean(**kwargs)
+                df = df.rolling(**kwargs).mean()
             elif rolling_type == 'std':
-                df = pd.rolling_std(**kwargs)
+                df = df.rolling(**kwargs).std()
             elif rolling_type == 'sum':
-                df = pd.rolling_sum(**kwargs)
+                df = df.rolling(**kwargs).sum()
         elif rolling_type == 'cumsum':
             df = df.cumsum()
         if min_periods: