def test_autoreg_info_criterion(lag):
data = sunspots.load(as_pandas=False)
endog = data.endog
endog_tmp = endog[16 - lag:]
r = AutoReg(endog_tmp, lags=lag).fit()
# See issue #324 for the corrections vs. R
k_ar = len(r.model.ar_lags)
k_trend = 1
log_sigma2 = np.log(r.sigma2)
aic = r.aic
aic = (aic - log_sigma2) * (1 + k_ar) / (1 + k_ar + k_trend)
aic += log_sigma2
hqic = r.hqic
hqic = (hqic - log_sigma2) * (1 + k_ar) / (1 + k_ar + k_trend)
hqic += log_sigma2
bic = r.bic
bic = (bic - log_sigma2) * (1 + k_ar) / (1 + k_ar + k_trend)
bic += log_sigma2
res1 = np.array([aic, hqic, bic, r.fpe])
# aic correction to match R
res2 = results_ar.ARLagResults("const").ic.T
assert_almost_equal(res1, res2[lag - 1, :], DECIMAL_6)
r2 = AutoReg(endog, lags=lag, hold_back=16).fit()
assert_allclose(r.aic, r2.aic)
assert_allclose(r.bic, r2.bic)
assert_allclose(r.hqic, r2.hqic)
assert_allclose(r.fpe, r2.fpe)
def test_ar_select_order_smoke():
data = sunspots.load(as_pandas=True).data["SUNACTIVITY"]
ar_select_order(data, 4, glob=True, trend="n")
ar_select_order(data, 4, glob=False, trend="n")
ar_select_order(data, 4, seasonal=True, period=12)
ar_select_order(data, 4, seasonal=False)
ar_select_order(data, 4, glob=True)
ar_select_order(data, 4, glob=True, seasonal=True, period=12)
def main():
"""
Toy program to test the KPSS autolag method of Hobijn et al (1998).
Unit tests using statsmodels data sets verified against SAS 9.3. To
use, modify the following lines in main KPSS method:
old:
if lags is None:
# from Kwiatkowski et al. referencing Schwert (1989)
lags = int(np.ceil(12. * np.power(nobs / 100., 1 / 4.)))
new:
if lags is None:
# autolag method of Hobijn et al. (1998)
lags = _kpss_autolag(resids, nobs)
"""
print("KPSS autolag method of Hobijn et al. (1998)")
# real GDP from macrodata data set
with warnings.catch_warnings(record=True) as w:
res = kpss(macrodata.load().data['realgdp'], 'c')
print(" realgdp('c'): stat =", "{0:0.5f}".format(res[0]), " pval =",
"{0:0.5f}".format(res[1]), " lags =", format(res[2]))
assert_almost_equal(res[0], 2.06851, decimal=3)
assert_equal(res[2], 9)
# sunspot activity from sunspots data set
with warnings.catch_warnings(record=True) as w:
res = kpss(sunspots.load().data['SUNACTIVITY'], 'c')
print(" sunactivity('c'): stat =", "{0:0.5f}".format(res[0]), " pval =",
"{0:0.5f}".format(res[1]), " lags =", format(res[2]))
assert_almost_equal(res[0], 0.66987, decimal=3)
assert_equal(res[2], 7)
# volumes from nile data set
with warnings.catch_warnings(record=True) as w:
res = kpss(nile.load().data['volume'], 'c')
print(" volume('c'): stat =", "{0:0.5f}".format(res[0]), " pval =",
"{0:0.5f}".format(res[1]), " lags =", format(res[2]))
assert_almost_equal(res[0], 0.86912, decimal=3)
assert_equal(res[2], 5)
# log-coinsurance from randhie data set
with warnings.catch_warnings(record=True) as w:
res = kpss(randhie.load().data['lncoins'], 'ct')
print(" lncoins('ct'): stat =", "{0:0.5f}".format(res[0]), " pval =",
"{0:0.5f}".format(res[1]), " lags =", format(res[2]))
assert_almost_equal(res[0], 0.36762, decimal=3)
assert_equal(res[2], 75)
# in-vehicle time from modechoice data set
with warnings.catch_warnings(record=True) as w:
res = kpss(modechoice.load().data['invt'], 'ct')
print(" invt('ct'): stat =", "{0:0.5f}".format(res[0]), " pval =",
"{0:0.5f}".format(res[1]), " lags =", format(res[2]))
assert_almost_equal(res[0], 0.40258, decimal=3)
assert_equal(res[2], 18)
def test_lags(self):
# real GDP from macrodata data set
with pytest.warns(InterpolationWarning):
res = kpss(self.x, 'c', nlags='auto')
assert_equal(res[2], 9)
# real interest rates from macrodata data set
res = kpss(sunspots.load(True).data['SUNACTIVITY'], 'c', nlags='auto')
assert_equal(res[2], 7)
# volumes from nile data set
with pytest.warns(InterpolationWarning):
res = kpss(nile.load(True).data['volume'], 'c', nlags='auto')
assert_equal(res[2], 5)
# log-coinsurance from randhie data set
with pytest.warns(InterpolationWarning):
res = kpss(randhie.load(True).data['lncoins'], 'ct', nlags='auto')
assert_equal(res[2], 75)
# in-vehicle time from modechoice data set
with pytest.warns(InterpolationWarning):
res = kpss(modechoice.load(True).data['invt'], 'ct', nlags='auto')
assert_equal(res[2], 18)
def test_lags(self):
# real GDP from macrodata data set
with pytest.warns(InterpolationWarning):
res = kpss(self.x, "c", nlags="auto")
assert_equal(res[2], 9)
# real interest rates from macrodata data set
res = kpss(sunspots.load().data["SUNACTIVITY"], "c", nlags="auto")
assert_equal(res[2], 7)
# volumes from nile data set
with pytest.warns(InterpolationWarning):
res = kpss(nile.load().data["volume"], "c", nlags="auto")
assert_equal(res[2], 5)
# log-coinsurance from randhie data set
with pytest.warns(InterpolationWarning):
res = kpss(randhie.load().data["lncoins"], "ct", nlags="auto")
assert_equal(res[2], 75)
# in-vehicle time from modechoice data set
with pytest.warns(InterpolationWarning):
res = kpss(modechoice.load().data["invt"], "ct", nlags="auto")
assert_equal(res[2], 18)
def test_lags(self):
# real GDP from macrodata data set
with warnings.catch_warnings(record=True):
lags = kpss(self.x, 'c', lags='auto')[2]
assert_equal(lags, 9)
# real interest rates from macrodata data set
with warnings.catch_warnings(record=True):
lags = kpss(sunspots.load().data['SUNACTIVITY'], 'c',
lags='auto')[2]
assert_equal(lags, 7)
# volumes from nile data set
with warnings.catch_warnings(record=True):
lags = kpss(nile.load().data['volume'], 'c', lags='auto')[2]
assert_equal(lags, 5)
# log-coinsurance from randhie data set
with warnings.catch_warnings(record=True):
lags = kpss(randhie.load().data['lncoins'], 'ct', lags='auto')[2]
assert_equal(lags, 75)
# in-vehicle time from modechoice data set
with warnings.catch_warnings(record=True):
lags = kpss(modechoice.load().data['invt'], 'ct', lags='auto')[2]
assert_equal(lags, 18)
def test_lags(self):
# real GDP from macrodata data set
with warnings.catch_warnings(record=True):
lags = kpss(self.x, 'c', lags='auto')[2]
assert_equal(lags, 9)
# real interest rates from macrodata data set
with warnings.catch_warnings(record=True):
lags = kpss(sunspots.load().data['SUNACTIVITY'], 'c',
lags='auto')[2]
assert_equal(lags, 7)
# volumes from nile data set
with warnings.catch_warnings(record=True):
lags = kpss(nile.load().data['volume'], 'c', lags='auto')[2]
assert_equal(lags, 5)
# log-coinsurance from randhie data set
with warnings.catch_warnings(record=True):
lags = kpss(randhie.load().data['lncoins'], 'ct', lags='auto')[2]
assert_equal(lags, 75)
# in-vehicle time from modechoice data set
with warnings.catch_warnings(record=True):
lags = kpss(modechoice.load().data['invt'], 'ct', lags='auto')[2]
assert_equal(lags, 18)
def test_autoreg_info_criterion(lag):
data = sunspots.load()
endog = np.asarray(data.endog)
endog_tmp = endog[16 - lag:]
r = AutoReg(endog_tmp, lags=lag).fit()
# See issue #324 for the corrections vs. R
aic = r.aic
hqic = r.hqic
bic = r.bic
res1 = np.array([aic, hqic, bic, r.fpe])
# aic correction to match R
res2 = results_ar.ARLagResults("const").ic.T
comp = res2[lag - 1, :].copy()
k = 2 + lag
pen = np.array([2, 2 * np.log(np.log(r.nobs)), np.log(r.nobs)])
comp[:3] = -2 * r.llf + pen * k
assert_almost_equal(res1, comp, DECIMAL_6)
r2 = AutoReg(endog, lags=lag, hold_back=16).fit()
assert_allclose(r.aic, r2.aic)
assert_allclose(r.bic, r2.bic)
assert_allclose(r.hqic, r2.hqic)
assert_allclose(r.fpe, r2.fpe)
def setup_class(cls):
data = sunspots.load(as_pandas=False)
cls.res1 = AutoReg(data.endog, lags=9, trend="n").fit()
cls.res2 = results_ar.ARResultsOLS(constant=False)
def setup_class(cls):
data = sunspots.load(as_pandas=True)
data.endog.index = list(range(len(data.endog)))
cls.res1 = AutoReg(data.endog, lags=9).fit()
cls.res2 = results_ar.ARResultsOLS(constant=True)
y = arma_generate_sample(ar, ma, nobs, 2)
y -= y.mean() #I have not checked treatment of mean yet, so remove
mod = MLEGLS(y)
mod.nar, mod.nma = 2, 2 #needs to be added, no init method
mod.nobs = len(y)
res = mod.fit(start_params=[0.1, -0.8, 0.2, 0.1, 1.])
print('DGP', ar, ma)
print(res.params)
from statsmodels.regression import yule_walker
print(yule_walker(y, 2))
#resi = mod.fit_invertible(start_params=[0.1,0,0.2,0, 0.5])
#print(resi.params
arpoly, mapoly = getpoly(mod, res.params[:-1])
data = sunspots.load()
#ys = data.endog[-100:]
## ys = data.endog[12:]-data.endog[:-12]
## ys -= ys.mean()
## mods = MLEGLS(ys)
## mods.nar, mods.nma = 13, 1 #needs to be added, no init method
## mods.nobs = len(ys)
## ress = mods.fit(start_params=np.r_[0.4, np.zeros(12), [0.2, 5.]],maxiter=200)
## print(ress.params
## import matplotlib.pyplot as plt
## plt.plot(data.endog[1])
## #plt.show()
sigma = mod._params2cov(res.params[:-1], nobs) * res.params[-1]**2
print(mvn_loglike(y, sigma))
llo = mvn_nloglike_obs(y, sigma)
def setup_class(cls):
data = sunspots.load()
cls.res1 = AutoReg(np.asarray(data.endog), lags=9, trend="n").fit()
cls.res2 = results_ar.ARResultsOLS(constant=False)
y = arma_generate_sample(ar, ma, nobs, 2)
y -= y.mean() #I have not checked treatment of mean yet, so remove
mod = MLEGLS(y)
mod.nar, mod.nma = 2, 2 #needs to be added, no init method
mod.nobs = len(y)
res = mod.fit(start_params=[0.1, -0.8, 0.2, 0.1, 1.])
print('DGP', ar, ma)
print(res.params)
from statsmodels.regression import yule_walker
print(yule_walker(y, 2))
#resi = mod.fit_invertible(start_params=[0.1,0,0.2,0, 0.5])
#print(resi.params
arpoly, mapoly = getpoly(mod, res.params[:-1])
data = sunspots.load(as_pandas=False)
#ys = data.endog[-100:]
## ys = data.endog[12:]-data.endog[:-12]
## ys -= ys.mean()
## mods = MLEGLS(ys)
## mods.nar, mods.nma = 13, 1 #needs to be added, no init method
## mods.nobs = len(ys)
## ress = mods.fit(start_params=np.r_[0.4, np.zeros(12), [0.2, 5.]],maxiter=200)
## print(ress.params
## import matplotlib.pyplot as plt
## plt.plot(data.endog[1])
## #plt.show()
sigma = mod._params2cov(res.params[:-1], nobs) * res.params[-1]**2
print(mvn_loglike(y, sigma))
llo = mvn_nloglike_obs(y, sigma)
cv_inf = mackinnoncrit()
assert np.all(cv_50 <= cv_inf)
def test_adf_short_timeseries():
# GH 262
import numpy as np
from arch.unitroot import ADF
x = np.asarray([0., 0., 0., 0., 0., 0., 1., 1., 0., 0.])
adf = ADF(x)
assert_almost_equal(adf.stat, -2.236, decimal=3)
assert adf.lags == 1
kpss_autolag_data = ((macrodata.load().data['realgdp'], 'c',
9), (sunspots.load().data['SUNACTIVITY'], 'c',
7), (nile.load().data['volume'], 'c',
5), (randhie.load().data['lncoins'], 'ct', 75),
(modechoice.load().data['invt'], 'ct', 18))
@pytest.mark.filterwarnings('ignore::DeprecationWarning')
@pytest.mark.parametrize('data,trend,lags', kpss_autolag_data)
def test_kpss_data_dependent_lags(data, trend, lags):
# real GDP from macrodata data set
kpss = KPSS(data, trend=trend)
assert_equal(kpss.lags, lags)
za_test_result = namedtuple('za_test_result', [
'stat',
cv_inf = mackinnoncrit()
assert np.all(cv_50 <= cv_inf)
def test_adf_short_timeseries():
# GH 262
import numpy as np
from arch.unitroot import ADF
x = np.asarray([0., 0., 0., 0., 0., 0., 1., 1., 0., 0.])
adf = ADF(x)
assert_almost_equal(adf.stat, -2.236, decimal=3)
assert adf.lags == 1
kpss_autolag_data = ((macrodata.load().data['realgdp'], 'c', 9),
(sunspots.load().data['SUNACTIVITY'], 'c', 7),
(nile.load().data['volume'], 'c', 5),
(randhie.load().data['lncoins'], 'ct', 75),
(modechoice.load().data['invt'], 'ct', 18))
@pytest.mark.filterwarnings('ignore::DeprecationWarning')
@pytest.mark.parametrize('data,trend,lags', kpss_autolag_data)
def test_kpss_data_dependent_lags(data, trend, lags):
# real GDP from macrodata data set
kpss = KPSS(data, trend=trend)
assert_equal(kpss.lags, lags)
za_test_result = namedtuple('za_test_result',
['stat', 'pvalue', 'lags', 'trend',
