add more comprehensive tests

ewstools/core.py

@@ -99,6 +99,7 @@ def __init__(self, data, transition=None):
             print(
                 "Make sure to provide data as either a list, np.ndarray or pd.Series\n"
             )
+            return
 
         # Set state and transition attributes
         self.state = df_state
@@ -1129,6 +1130,7 @@ def __init__(self, data, transition=None):
         if type(data) != pd.DataFrame:
             print("\nERROR: data has been provided as type {}".format(type(data)))
             print("Please provide data as a pandas DataFrame.\n")
+            return
 
         # Set state and transition attributes
         self.state = data
@@ -1224,7 +1226,8 @@ def compute_covar(self, rolling_window=0.25, leading_eval=False):
         If residuals have not been computed, computation will be
         performed over state variable.
 
-        Put into 'ews' dataframe
+        Put covariance matrices into self.covar
+        Put leading eigenvalue into self.ews
 
         Parameters
         ----------
@@ -1282,6 +1285,55 @@ def compute_covar(self, rolling_window=0.25, leading_eval=False):
             series_evals = pd.Series(ar_evals, index=df_pre.index)
             self.ews["covar_leading_eval"] = series_evals
 
+    def compute_corr(self, rolling_window=0.25):
+        """
+        Compute the (Pearson) correlation matrix over a rolling window.
+        If residuals have not been computed, computation will be
+        performed over state variable.
+
+        Put correlation matrices into self.corr
+
+        Parameters
+        ----------
+        rolling_window : float
+            Length of rolling window used to compute variance. Can be specified
+            as an absolute value or as a proportion of the length of the
+            data being analysed. Default is 0.25.
+
+        Returns
+        -------
+        None.
+
+        """
+
+        # Get time series data prior to transition
+        if self.transition:
+            df_pre = self.state[self.state.index <= self.transition]
+        else:
+            df_pre = self.state
+
+        # Get absolute size of rollling window if given as a proportion
+        if 0 < rolling_window <= 1:
+            rw_absolute = int(rolling_window * len(df_pre))
+        else:
+            rw_absolute = rolling_window
+
+        # If residuals column exists, compute over residuals.
+        if "{}_residuals".format(self.var_names[0]) in df_pre.columns:
+            col_names_to_compute = [
+                "{}_residuals".format(var) for var in self.var_names
+            ]
+        else:
+            col_names_to_compute = self.var_names
+
+        # Compute correlation matrix
+        df_corr = (
+            df_pre[col_names_to_compute]
+            .rolling(window=rw_absolute)
+            .corr(method="Pearson")
+        )
+        self.corr = df_corr
+
 
 # -----------------------------
 # Eigenvalue reconstruction

tests/test_ewstools.py

@@ -60,12 +60,22 @@ def test_MultiTimeSeries_ews():
     df.index.name = "index_name"
 
     # Create MultiTimeSeries object
-    mts = ewstools.core.MultiTimeSeries(df, transition=8)
+    mts = ewstools.MultiTimeSeries(np.array([1, 2, 3]))  # invalid entry
+    mts = ewstools.MultiTimeSeries(df, transition=8)
+    mts.detrend(method="XXX", bandwidth=0.2)  # invalid detrend method
     mts.detrend(method="Gaussian", bandwidth=0.2)
+    mts.detrend(method="Gaussian", bandwidth=20)
+    mts.detrend(method="Lowess", span=0.2)
+    mts.detrend(method="Lowess", span=20)
+
     mts.compute_covar(rolling_window=0.25, leading_eval=True)
+    mts.compute_covar(rolling_window=20, leading_eval=False)
+    mts.compute_corr(rolling_window=0.25)
+    mts.compute_corr(rolling_window=20)
 
     assert type(mts.ews) == pd.DataFrame
     assert type(mts.covar) == pd.DataFrame
+    assert type(mts.corr) == pd.DataFrame
     assert "x_residuals" in mts.state.columns
     assert "z_smoothing" in mts.state.columns
     assert "covar_leading_eval" in mts.ews.columns
@@ -81,6 +91,7 @@ def test_TimeSeries_init():
     xVals = 5 + np.random.normal(0, 1, len(tVals))
 
     # Create TimeSeries object using np.ndarray
+    ts = ewstools.TimeSeries("hello")  # invalid entry
     ts = ewstools.TimeSeries(xVals)
     assert type(ts.state) == pd.DataFrame
     assert type(ts.ews) == pd.DataFrame