Merge 7512194 into e1be28e

statsmodels/compatnp/py3k.py

@@ -34,6 +34,7 @@ def isfileobj(f):
     def open_latin1(filename, mode='r'):
         return open(filename, mode=mode, encoding='iso-8859-1')
     strchar = 'U'
+    string_types = basestring
     from io import BytesIO, StringIO  #statsmodels
 else:
     bytes = str
@@ -52,6 +53,7 @@ def open_latin1(filename, mode='r'):
         return open(filename, mode=mode)
     from StringIO import StringIO
     BytesIO = StringIO
+    string_types = str
 
 def getexception():
     return sys.exc_info()[1]

statsmodels/sandbox/stats/diagnostic.py

@@ -34,7 +34,8 @@
 from scipy import stats
 from statsmodels.regression.linear_model import OLS
 from statsmodels.tools.tools import add_constant
-from statsmodels.tsa.stattools import acf, adfuller
+from statsmodels.tsa.stattools import acf
+from statsmodels.tsa.unitroot import adfuller
 from statsmodels.tsa.tsatools import lagmat
 
 #get the old signature back so the examples work

statsmodels/tsa/cointegration.py

@@ -0,0 +1,70 @@
+"""
+Estimation and testing of cointegrated time series
+"""
+from __future__ import division
+
+import numpy as np
+
+from statsmodels.regression.linear_model import OLS
+from statsmodels.tools.tools import add_constant
+from statsmodels.tsa.unitroot import mackinnonp, mackinnoncrit
+
+
+def coint(y1, y2, regression="c"):
+    """
+    This is a simple cointegration test. Uses unit-root test on residuals to
+    test for cointegrated relationship
+
+    See Hamilton (1994) 19.2
+
+    Parameters
+    ----------
+    y1 : array_like, 1d
+        first element in cointegrating vector
+    y2 : array_like
+        remaining elements in cointegrating vector
+    c : str {'c'}
+        Included in regression
+        * 'c' : Constant
+
+    Returns
+    -------
+    coint_t : float
+        t-statistic of unit-root test on residuals
+    pvalue : float
+        MacKinnon's approximate p-value based on MacKinnon (1994)
+    crit_value : dict
+        Critical values for the test statistic at the 1 %, 5 %, and 10 %
+        levels.
+
+    Notes
+    -----
+    The Null hypothesis is that there is no cointegration, the alternative
+    hypothesis is that there is cointegrating relationship. If the pvalue is
+    small, below a critical size, then we can reject the hypothesis that there
+    is no cointegrating relationship.
+
+    P-values are obtained through regression surface approximation from
+    MacKinnon 1994.
+
+    References
+    ----------
+    MacKinnon, J.G. 1994.  "Approximate asymptotic distribution functions for
+        unit-root and cointegration tests.  `Journal of Business and Economic
+        Statistics` 12, 167-76.
+
+    """
+    regression = regression.lower()
+    if regression not in ['c', 'nc', 'ct', 'ctt']:
+        raise ValueError("regression option %s not understood") % regression
+    y1 = np.asarray(y1)
+    y2 = np.asarray(y2)
+    if regression == 'c':
+        y2 = add_constant(y2, prepend=False)
+    st1_resid = OLS(y1, y2).fit().resid  # stage one residuals
+    lgresid_cons = add_constant(st1_resid[0:-1], prepend=False)
+    uroot_reg = OLS(st1_resid[1:], lgresid_cons).fit()
+    coint_t = (uroot_reg.params[0] - 1) / uroot_reg.bse[0]
+    pvalue = mackinnonp(coint_t, regression="c", num_unit_roots=2)
+    crit_value = mackinnoncrit(num_unit_roots=1, regression="c", nobs=len(y1))
+    return coint_t, pvalue, crit_value

statsmodels/tsa/critical_values/__init__.py

@@ -0,0 +1 @@
+__author__ = 'kevin.sheppard'

statsmodels/tsa/critical_values/dfgls.py

@@ -0,0 +1,34 @@
+"""
+Contains values used to approximate the critical value and
+p-value from DFGLS statistics
+
+These have been computed using the methodology of MacKinnon (1994) and (2010).
+simulation.
+"""
+
+from numpy import array
+
+dfgls_cv_approx = {'c': array([[-2.56781793e+00, -2.05575392e+01, 1.82727674e+02,
+                                -1.77866664e+03],
+                               [-1.94363325e+00, -2.17272746e+01, 2.60815068e+02,
+                                -2.26914916e+03],
+                               [-1.61998241e+00, -2.32734708e+01, 3.06474378e+02,
+                                -2.57483557e+03]]),
+                   'ct': array([[-3.40689134, -21.69971242, 27.26295939, -816.84404772],
+                                [-2.84677178, -19.69109364, 84.7664136, -799.40722401],
+                                [-2.55890707, -19.42621991, 116.53759752, -840.31342847]])}
+
+dfgls_tau_max = {'c': 13.365361509140614,
+                 'ct': 8.73743383728356}
+
+dfgls_tau_min = {'c': -17.561302895074206,
+                 'ct': -13.681153542634465}
+
+dfgls_tau_star = {'c': -0.4795076091714674,
+                  'ct': -2.1960404365401298}
+
+dfgls_large_p = {'c': array([0.50612497, 0.98305664, -0.05648525, 0.00140875]),
+                 'ct': array([2.60561421, 1.67850224, 0.0373599, -0.01017936])}
+
+dfgls_small_p = {'c': array([0.67422739, 1.25475826, 0.03572509]),
+                 'ct': array([2.38767685, 1.57454737, 0.05754439])}