- index_rolling_corr is a new additional methods to compare ETFs with index (benchmarks) in AssetList

- bug fixed in tracking_difference_annualized
- howto jupyter notebook for index funds

Author: chilango74

main.py

@@ -1,6 +1,6 @@
 import okama as ok
 
-ls4 = ['SPY.US', 'BND.US', 'GLD.US', 'VNQ.US']
-curr = 'USD'
-four_assets = ok.EfficientFrontier(symbols=ls4, curr=curr, n_points=100)
-ok.Plots(ls4, curr=curr).plot_pair_ef()
+funds1 = ['MCFTR.INDX', '0177-71671092.PIF', '0890-94127385.PIF']
+curr = 'RUB'
+x1 = ok.AssetList(funds1, curr=curr)
+x1.tracking_difference_annualized

notebooks/Efficient Frontier.ipynb

@@ -302,7 +302,7 @@
    "cell_type": "markdown",
    "metadata": {},
    "source": [
-    "Lets add a popular fisical gold and real estate ETFs..."
+    "Let's add a popular fisical gold and real estate ETFs..."
    ]
   },
   {

notebooks/index funds perfomance.ipynb

@@ -598,12 +598,21 @@ def index_corr(self):
         """
         return Index.cov_cor(self.ror, fn='corr')
 
+    def index_rolling_corr(self, window=60):
+        """
+        Returns the rolling correlation with the index (or benchmark) time series for the assets.
+        Index should be in the first position (first column).
+        The period should be at least 12 months.
+        window - the rolling window size (default is 5 years).
+        """
+        return Index.rolling_cov_cor(self.ror, window=window, fn='corr')
+
     @property
     def index_beta(self):
         """
         Returns beta coefficient time series for the assets.
         Index (or benchmark) should be in the first position (first column).
-        The period should be at least 12 months.
+        Rolling window size should be at least 12 months.
         """
         return Index.beta(self.ror)
 

okama/assets.py

@@ -366,7 +366,8 @@ def tracking_difference_annualized(tracking_diff: pd.DataFrame) -> pd.DataFrame:
         """
         pwr = 12 / (1. + np.arange(tracking_diff.shape[0]))
         y = abs(tracking_diff)
-        diff = np.sign(tracking_diff) * (y + 1.).pow(pwr, axis=0) - 1.
+        diff = (y + 1.).pow(pwr, axis=0) - 1.
+        diff = np.sign(tracking_diff) * diff
         return diff.iloc[_MONTHS_PER_YEAR - 1:]  # returns for the first 11 months can't be annualized
 
     @staticmethod
@@ -392,12 +393,32 @@ def cov_cor(ror: pd.DataFrame, fn: str) -> pd.DataFrame:
         if ror.shape[1] < 2:
             raise ValueError('At least 2 symbols should be provided.')
         if fn not in ['cov', 'corr']:
-            raise ValueError('fn should be cor or cov')
+            raise ValueError('fn should be corr or cov')
         cov_matrix_ts = getattr(ror.expanding(), fn)()
         cov_matrix_ts = cov_matrix_ts.drop(index=ror.columns[1:], level=1).droplevel(1)
         cov_matrix_ts.drop(columns=ror.columns[0], inplace=True)
         return cov_matrix_ts.iloc[_MONTHS_PER_YEAR:]
 
+    @staticmethod
+    def rolling_cov_cor(ror: pd.DataFrame, window: int = 60, fn: str = 'corr') -> pd.DataFrame:
+        """
+        Returns the rolling correlation (or covariance) time series.
+        The period should be at least 12 months.
+        """
+        if ror.shape[1] < 2:
+            raise ValueError('At least 2 symbols should be provided.')
+        if fn not in ['cov', 'corr']:
+            raise ValueError('fn should be corr or cov')
+        if window < _MONTHS_PER_YEAR:
+            raise ValueError('window size should be at least 12 months')
+        if not isinstance(window, int):
+            raise ValueError('window should be an integer')
+        cov_matrix_ts = getattr(ror.rolling(window=window), fn)()
+        cov_matrix_ts = cov_matrix_ts.drop(index=ror.columns[1:], level=1).droplevel(1)
+        cov_matrix_ts.drop(columns=ror.columns[0], inplace=True)
+        cov_matrix_ts.dropna(inplace=True)
+        return cov_matrix_ts
+
     @staticmethod
     def beta(ror: pd.DataFrame) -> pd.DataFrame:
         """