def test_pickle(self):
from statsmodels.compatnp.py3k import BytesIO
fh = BytesIO()
#test wrapped results load save pickle
self.res.save(fh)
fh.seek(0, 0)
res_unpickled = self.res.__class__.load(fh)
assert_(type(res_unpickled) is type(self.res))
def test_pickle(self):
from statsmodels.compatnp.py3k import BytesIO
fh = BytesIO()
#test wrapped results load save pickle
self.res1.save(fh)
fh.seek(0,0)
res_unpickled = self.res1.__class__.load(fh)
assert_(type(res_unpickled) is type(self.res1))
def test_stata_writer_array():
buf = BytesIO()
dta = macrodata.load().data
dta = DataFrame.from_records(dta)
dta.columns = ["v%d" % i for i in range(1,15)]
writer = StataWriter(buf, dta.values)
writer.write_file()
buf.seek(0)
dta2 = genfromdta(buf)
dta = dta.to_records(index=False)
assert_array_equal(dta, dta2)
def test_stata_writer_structured():
buf = BytesIO()
dta = macrodata.load().data
dtype = dta.dtype
dta = dta.astype(np.dtype([('year', int),
('quarter', int)] + dtype.descr[2:]))
writer = StataWriter(buf, dta)
writer.write_file()
buf.seek(0)
dta2 = genfromdta(buf)
assert_array_equal(dta, dta2)
def test_stata_writer_pandas():
buf = BytesIO()
dta = macrodata.load().data
dtype = dta.dtype
#as of 0.9.0 pandas only supports i8 and f8
dta = dta.astype(np.dtype([('year', 'i8'),
('quarter', 'i8')] + dtype.descr[2:]))
dta = DataFrame.from_records(dta)
writer = StataWriter(buf, dta)
writer.write_file()
buf.seek(0)
dta2 = genfromdta(buf)
ptesting.assert_frame_equal(dta.reset_index(), DataFrame.from_records(dta2))
def test_missing_roundtrip():
buf = BytesIO()
dta = np.array([(np.nan, np.inf, "")],
dtype=[("double_miss", float), ("float_miss", np.float32),
("string_miss", "a1")])
writer = StataWriter(buf, dta)
writer.write_file()
buf.seek(0)
dta = genfromdta(buf, missing_flt=np.nan)
assert_(isnull(dta[0][0]))
assert_(isnull(dta[0][1]))
assert_(dta[0][2] == asbytes(""))
dta = genfromdta(os.path.join(curdir, "results/data_missing.dta"),
missing_flt=-999)
assert_(np.all([dta[0][i] == -999 for i in range(5)]))
def check_pickle(obj):
from statsmodels.compatnp.py3k import BytesIO
fh = BytesIO()
pickle.dump(obj, fh)
plen = fh.tell()
fh.seek(0, 0)
res = pickle.load(fh)
fh.close()
return res, plen
def test_pickle_wrapper(self):
from statsmodels.iolib.smpickle import save_pickle, load_pickle
from statsmodels.compatnp.py3k import BytesIO
fh = BytesIO() #use cPickle with binary content
#test unwrapped results load save pickle
self.results._results.save(fh)
fh.seek(0, 0)
res_unpickled = self.results._results.__class__.load(fh)
assert_(type(res_unpickled) is type(self.results._results))
#test wrapped results load save
fh.seek(0, 0)
#save_pickle(self.results, fh)
self.results.save(fh)
fh.seek(0, 0)
#res_unpickled = load_pickle(fh)
res_unpickled = self.results.__class__.load(fh)
fh.close()
#print type(res_unpickled)
assert_(type(res_unpickled) is type(self.results))
before = sorted(self.results.__dict__.keys())
after = sorted(res_unpickled.__dict__.keys())
assert_(before == after, msg='not equal %r and %r' % (before, after))
before = sorted(self.results._results.__dict__.keys())
after = sorted(res_unpickled._results.__dict__.keys())
assert_(before == after, msg='not equal %r and %r' % (before, after))
before = sorted(self.results.model.__dict__.keys())
after = sorted(res_unpickled.model.__dict__.keys())
assert_(before == after, msg='not equal %r and %r' % (before, after))
before = sorted(self.results._cache.keys())
after = sorted(res_unpickled._cache.keys())
assert_(before == after, msg='not equal %r and %r' % (before, after))
def test_pickle():
import tempfile
from numpy.testing import assert_equal
tmpdir = tempfile.mkdtemp(prefix='pickle')
a = range(10)
save_pickle(a, tmpdir+'/res.pkl')
b = load_pickle(tmpdir+'/res.pkl')
assert_equal(a, b)
#cleanup, tested on Windows
try:
import os
os.remove(tmpdir+'/res.pkl')
os.rmdir(tmpdir)
except (OSError, IOError):
pass
assert not os.path.exists(tmpdir)
#test with file handle
from statsmodels.compatnp.py3k import BytesIO
fh = BytesIO()
save_pickle(a, fh)
fh.seek(0,0)
c = load_pickle(fh)
fh.close()
assert_equal(a,b)
def check_pickle(obj):
from statsmodels.compatnp.py3k import BytesIO
fh = BytesIO()
pickle.dump(obj, fh)
plen = fh.tell()
fh.seek(0, 0)
res = pickle.load(fh)
fh.close()
return res, plen
def test_stata_writer_pandas():
buf = BytesIO()
dta = macrodata.load().data
dtype = dta.dtype
#as of 0.9.0 pandas only supports i8 and f8
dta = dta.astype(
np.dtype([('year', 'i8'), ('quarter', 'i8')] + dtype.descr[2:]))
dta4 = dta.astype(
np.dtype([('year', 'i4'), ('quarter', 'i4')] + dtype.descr[2:]))
dta = DataFrame.from_records(dta)
dta4 = DataFrame.from_records(dta4)
# dta is int64 'i8' given to Stata writer
writer = StataWriter(buf, dta)
writer.write_file()
buf.seek(0)
dta2 = genfromdta(buf)
dta5 = DataFrame.from_records(dta2)
# dta2 is int32 'i4' returned from Stata reader
if dta5.dtypes[1] is np.dtype('int64'):
ptesting.assert_frame_equal(dta.reset_index(), dta5)
else:
# don't check index because it has different size, int32 versus int64
ptesting.assert_frame_equal(dta4, dta5[dta5.columns[1:]])
def test_stata_writer_pandas():
buf = BytesIO()
dta = macrodata.load().data
dtype = dta.dtype
#as of 0.9.0 pandas only supports i8 and f8
dta = dta.astype(np.dtype([('year', 'i8'),
('quarter', 'i8')] + dtype.descr[2:]))
dta4 = dta.astype(np.dtype([('year', 'i4'),
('quarter', 'i4')] + dtype.descr[2:]))
dta = DataFrame.from_records(dta)
dta4 = DataFrame.from_records(dta4)
# dta is int64 'i8' given to Stata writer
writer = StataWriter(buf, dta)
writer.write_file()
buf.seek(0)
dta2 = genfromdta(buf)
dta5 = DataFrame.from_records(dta2)
# dta2 is int32 'i4' returned from Stata reader
if dta5.dtypes[1] is np.dtype('int64'):
ptesting.assert_frame_equal(dta.reset_index(), dta5)
else:
# don't check index because it has different size, int32 versus int64
ptesting.assert_frame_equal(dta4, dta5[dta5.columns[1:]])
def test_pickle_wrapper(self):
from statsmodels.iolib.smpickle import save_pickle, load_pickle
from statsmodels.compatnp.py3k import BytesIO
fh = BytesIO() #use cPickle with binary content
#test unwrapped results load save pickle
self.results._results.save(fh)
fh.seek(0,0)
res_unpickled = self.results._results.__class__.load(fh)
assert_(type(res_unpickled) is type(self.results._results))
#test wrapped results load save
fh.seek(0,0)
#save_pickle(self.results, fh)
self.results.save(fh)
fh.seek(0,0)
#res_unpickled = load_pickle(fh)
res_unpickled = self.results.__class__.load(fh)
fh.close()
#print type(res_unpickled)
assert_(type(res_unpickled) is type(self.results))
before = sorted(self.results.__dict__.keys())
after = sorted(res_unpickled.__dict__.keys())
assert_(before == after, msg='not equal %r and %r' % (before, after))
before = sorted(self.results._results.__dict__.keys())
after = sorted(res_unpickled._results.__dict__.keys())
assert_(before == after, msg='not equal %r and %r' % (before, after))
before = sorted(self.results.model.__dict__.keys())
after = sorted(res_unpickled.model.__dict__.keys())
assert_(before == after, msg='not equal %r and %r' % (before, after))
before = sorted(self.results._cache.keys())
after = sorted(res_unpickled._cache.keys())
assert_(before == after, msg='not equal %r and %r' % (before, after))
def test_datetime_roundtrip():
dta = np.array([(1, datetime(2010, 1, 1), 2), (2, datetime(2010, 2, 1), 3),
(4, datetime(2010, 3, 1), 5)],
dtype=[('var1', float), ('var2', object), ('var3', float)])
buf = BytesIO()
writer = StataWriter(buf, dta, {"var2": "tm"})
writer.write_file()
buf.seek(0)
dta2 = genfromdta(buf)
assert_equal(dta, dta2)
dta = DataFrame.from_records(dta)
buf = BytesIO()
writer = StataWriter(buf, dta, {"var2": "tm"})
writer.write_file()
buf.seek(0)
dta2 = genfromdta(buf, pandas=True)
ptesting.assert_frame_equal(dta, dta2.drop('index', axis=1))
def test_datetime_roundtrip():
dta = np.array([(1, datetime(2010, 1, 1), 2),
(2, datetime(2010, 2, 1), 3),
(4, datetime(2010, 3, 1), 5)],
dtype=[('var1', float), ('var2', object), ('var3', float)])
buf = BytesIO()
writer = StataWriter(buf, dta, {"var2" : "tm"})
writer.write_file()
buf.seek(0)
dta2 = genfromdta(buf)
assert_equal(dta, dta2)
dta = DataFrame.from_records(dta)
buf = BytesIO()
writer = StataWriter(buf, dta, {"var2" : "tm"})
writer.write_file()
buf.seek(0)
dta2 = genfromdta(buf, pandas=True)
ptesting.assert_frame_equal(dta, dta2.drop('index', axis=1))
ss5 = '''\
2 - 3\t4.340\t0.691\t7.989\t***
2 - 1\t4.600\t0.951\t8.249\t***
3 - 2\t-4.340\t-7.989\t-0.691\t***
3 - 1\t0.260\t-3.389\t3.909\t-
1 - 2\t-4.600\t-8.249\t-0.951\t***
1 - 3\t-0.260\t-3.909\t3.389\t'''
#accommodate recfromtxt for python 3.2, requires bytes
ss = asbytes(ss)
ss2 = asbytes(ss2)
ss3 = asbytes(ss3)
ss5 = asbytes(ss5)
dta = np.recfromtxt(BytesIO(ss), names=("Rust", "Brand", "Replication"))
dta2 = np.recfromtxt(BytesIO(ss2),
names=("idx", "Treatment", "StressReduction"))
dta3 = np.recfromtxt(BytesIO(ss3), names=("Brand", "Relief"))
dta5 = np.recfromtxt(BytesIO(ss5),
names=('pair', 'mean', 'lower', 'upper', 'sig'),
delimiter='\t')
sas_ = dta5[[1, 3, 2]]
from statsmodels.stats.multicomp import (tukeyhsd, pairwise_tukeyhsd,
MultiComparison)
#import statsmodels.sandbox.stats.multicomp as multi
#print tukeyhsd(dta['Brand'], dta['Rust'])
def get_thsd(mci, alpha=0.05):
ss5 = '''\
2 - 3 4.340 0.691 7.989 ***
2 - 1 4.600 0.951 8.249 ***
3 - 2 -4.340 -7.989 -0.691 ***
3 - 1 0.260 -3.389 3.909 -
1 - 2 -4.600 -8.249 -0.951 ***
1 - 3 -0.260 -3.909 3.389 '''
#accommodate recfromtxt for python 3.2, requires bytes
ss = asbytes(ss)
ss2 = asbytes(ss2)
ss3 = asbytes(ss3)
ss5 = asbytes(ss5)
dta = np.recfromtxt(BytesIO(ss), names=("Rust","Brand","Replication"))
dta2 = np.recfromtxt(BytesIO(ss2), names = ("idx", "Treatment", "StressReduction"))
dta3 = np.recfromtxt(BytesIO(ss3), names = ("Brand", "Relief"))
dta5 = np.recfromtxt(BytesIO(ss5), names = ('pair', 'mean', 'lower', 'upper', 'sig'), delimiter='\t')
sas_ = dta5[[1,3,2]]
from statsmodels.stats.multicomp import (tukeyhsd, pairwise_tukeyhsd,
MultiComparison)
#import statsmodels.sandbox.stats.multicomp as multi
#print tukeyhsd(dta['Brand'], dta['Rust'])
def get_thsd(mci, alpha=0.05):
var_ = np.var(mci.groupstats.groupdemean(), ddof=len(mci.groupsunique))
means = mci.groupstats.groupmean
nobs = mci.groupstats.groupnobs
resi = tukeyhsd(means, nobs, var_, df=None, alpha=alpha,