From cd3a63415c5f89a986fd87c85a282d6bd82e67a6 Mon Sep 17 00:00:00 2001
From: Eddie Hebert <ehebert@quantopian.com>
Date: Wed, 25 Sep 2013 11:25:57 -0400
Subject: [PATCH] MAINT: Use pandas for volatility in risk metrics.

Continue on path of converting values stored inside of risk metrics
to use a DataFrame instead of storing multiple lists.

Also, the need for latest_dt in getting the current volatility for
the sharpe calculation, shows that we need to set the lastest_dt at
the beginning of the update loop.
---
 tests/test_events_through_risk.py  | 12 +++++++-----
 zipline/finance/risk/cumulative.py | 30 ++++++++++++++++--------------
 2 files changed, 23 insertions(+), 19 deletions(-)

diff --git a/tests/test_events_through_risk.py b/tests/test_events_through_risk.py
index 7a8b42b0..891ee22b 100644
--- a/tests/test_events_through_risk.py
+++ b/tests/test_events_through_risk.py
@@ -150,18 +150,19 @@ class TestEventsThroughRisk(unittest.TestCase):
         }
 
         for bar in gen:
-            current_dt = algo.get_datetime()
+            current_dt = algo.datetime
             crm = algo.perf_tracker.cumulative_risk_metrics
 
             np.testing.assert_almost_equal(
+                crm.algorithm_returns[current_dt],
                 expected_algorithm_returns[current_dt],
-                crm.algorithm_returns[-1],
                 decimal=6)
 
             np.testing.assert_almost_equal(
-                expected_sharpe[current_dt],
                 crm.metrics.sharpe[current_dt],
-                decimal=6)
+                expected_sharpe[current_dt],
+                decimal=6,
+                err_msg="Mismatch at %s" % (current_dt,))
 
     def test_minute_buy_and_hold(self):
         with trading.TradingEnvironment():
@@ -305,7 +306,8 @@ class TestEventsThroughRisk(unittest.TestCase):
                              "be one position after the first day.")
 
             self.assertTrue(
-                np.isnan(crm.algorithm_volatility[-1]),
+                np.isnan(
+                    crm.metrics.algorithm_volatility[algo.datetime.date()]),
                 "On the first day algorithm volatility does not exist.")
 
             second_msg = gen.next()
diff --git a/zipline/finance/risk/cumulative.py b/zipline/finance/risk/cumulative.py
index 507dad98..73aada7f 100644
--- a/zipline/finance/risk/cumulative.py
+++ b/zipline/finance/risk/cumulative.py
@@ -44,7 +44,9 @@ class RiskMetricsCumulative(object):
     METRIC_NAMES = (
         'alpha',
         'beta',
-        'sharpe'
+        'sharpe',
+        'algorithm_volatility',
+        'benchmark_volatility',
     )
 
     def __init__(self, sim_params, returns_frequency=None):
@@ -93,8 +95,6 @@ class RiskMetricsCumulative(object):
 
         self.compounded_log_returns = []
 
-        self.algorithm_volatility = []
-        self.benchmark_volatility = []
         self.algorithm_period_returns = []
         self.benchmark_period_returns = []
 
@@ -122,6 +122,10 @@ class RiskMetricsCumulative(object):
         return self.algorithm_returns.index[-1]
 
     def update(self, dt, algorithm_returns, benchmark_returns):
+        # Keep track of latest dt for use in to_dict and other methods
+        # that report current state.
+        self.latest_dt = dt
+
         self.algorithm_returns_cont[dt] = algorithm_returns
         self.algorithm_returns = self.algorithm_returns_cont.valid()
 
@@ -152,10 +156,10 @@ algorithm_returns ({algo_count}) in range {start} : {end} on {dt}"
             raise Exception(message)
 
         self.update_current_max()
-        self.benchmark_volatility.append(
-            self.calculate_volatility(self.benchmark_returns))
-        self.algorithm_volatility.append(
-            self.calculate_volatility(self.algorithm_returns))
+        self.metrics.benchmark_volatility[dt] = \
+            self.calculate_volatility(self.benchmark_returns)
+        self.metrics.algorithm_volatility[dt] = \
+            self.calculate_volatility(self.algorithm_returns)
 
         # caching the treasury rates for the minutely case is a
         # big speedup, because it avoids searching the treasury
@@ -181,10 +185,6 @@ algorithm_returns ({algo_count}) in range {start} : {end} on {dt}"
         self.information.append(self.calculate_information())
         self.max_drawdown = self.calculate_max_drawdown()
 
-        # Keep track of latest dt for use in to_dict and other methods
-        # that report current state.
-        self.latest_dt = dt
-
     def to_dict(self):
         """
         Creates a dictionary representing the state of the risk report.
@@ -193,8 +193,10 @@ algorithm_returns ({algo_count}) in range {start} : {end} on {dt}"
         period_label = self.last_return_date.strftime("%Y-%m")
         rval = {
             'trading_days': len(self.algorithm_returns.valid()),
-            'benchmark_volatility': self.benchmark_volatility[-1],
-            'algo_volatility': self.algorithm_volatility[-1],
+            'benchmark_volatility':
+            self.metrics.benchmark_volatility[self.latest_dt],
+            'algo_volatility':
+            self.metrics.algorithm_volatility[self.latest_dt],
             'treasury_period_return': self.treasury_period_return,
             'algorithm_period_return': self.algorithm_period_returns[-1],
             'benchmark_period_return': self.benchmark_period_returns[-1],
@@ -289,7 +291,7 @@ algorithm_returns ({algo_count}) in range {start} : {end} on {dt}"
         """
         http://en.wikipedia.org/wiki/Sharpe_ratio
         """
-        return sharpe_ratio(self.algorithm_volatility[-1],
+        return sharpe_ratio(self.metrics.algorithm_volatility[self.latest_dt],
                             self.algorithm_period_returns[-1],
                             self.treasury_period_return)