def Test_adjust_hk(self):
"""
Test Difference between Hong Kong China Adjustment using a Random Sample
random sample: stata_wtf90_hk_china_adjust_sitc4_check_sample.csv
"""
OPTIONS = {
'A': {
'dropAX': False,
'sitcr2': False,
'drop_nonsitcr2': False,
'adjust_hk': (False, None),
'intertemp_cntrycode': False,
'drop_incp_cntrycode': False,
'adjust_units': False,
'source_institution': 'un',
},
'B': {
'dropAX': False,
'sitcr2': False,
'drop_nonsitcr2': False,
'adjust_hk': (True, self.hkchina_rawdata),
'intertemp_cntrycode': False,
'drop_incp_cntrycode': False,
'adjust_units': False,
'source_institution': 'un',
},
}
#-SITC Level 4-#
A = construct_sitcr2(self.rawdata,
data_type='trade',
level=4,
**OPTIONS['A'])
B = construct_sitcr2(self.rawdata,
data_type='trade',
level=4,
**OPTIONS['B'])
RAW = pd.read_csv(TEST_DATA_DIR +
"stata_wtf90_hk_china_adjust_sitc4_check_sample.csv")
for idx, row in RAW.iterrows():
A_VAL = A.loc[(A.year == row.year) & (A.iiso3c == row.iiso3c) &
(A.eiso3c == row.eiso3c) &
(A.sitc4 == str(row.sitc4))]
if len(A_VAL) == 0:
print "Data not contained in A"
continue
A_VAL = A_VAL.get_value(index=A_VAL.index[0], col='value')
assert A_VAL == row.value
B_VAL = B.loc[(B.year == row.year) & (B.iiso3c == row.iiso3c) &
(B.eiso3c == row.eiso3c) &
(B.sitc4 == str(row.sitc4))]
if len(B_VAL) == 0:
print "Data not contained in B"
continue
B_VAL = B_VAL.get_value(index=B_VAL.index[0], col='value')
assert B_VAL == row.value_adj
def Test_adjust_hk(self):
"""
Test Difference between Hong Kong China Adjustment using a Random Sample
random sample: stata_wtf90_hk_china_adjust_sitc4_check_sample.csv
"""
OPTIONS = {
'A' : { 'dropAX' : False,
'sitcr2' : False,
'drop_nonsitcr2' : False,
'adjust_hk' : (False, None),
'intertemp_cntrycode' : False,
'drop_incp_cntrycode' : False,
'adjust_units' : False,
'source_institution' : 'un',
},
'B' : { 'dropAX' : False,
'sitcr2' : False,
'drop_nonsitcr2' : False,
'adjust_hk' : (True, self.hkchina_rawdata),
'intertemp_cntrycode' : False,
'drop_incp_cntrycode' : False,
'adjust_units' : False,
'source_institution' : 'un',
},
}
#-SITC Level 4-#
A = construct_sitcr2(self.rawdata, data_type='trade', level=4, **OPTIONS['A'])
B = construct_sitcr2(self.rawdata, data_type='trade', level=4, **OPTIONS['B'])
RAW = pd.read_csv(TEST_DATA_DIR + "stata_wtf90_hk_china_adjust_sitc4_check_sample.csv")
for idx,row in RAW.iterrows():
A_VAL = A.loc[(A.year == row.year) & (A.iiso3c == row.iiso3c) & (A.eiso3c == row.eiso3c) & (A.sitc4 == str(row.sitc4))]
if len(A_VAL) == 0:
print "Data not contained in A"
continue
A_VAL = A_VAL.get_value(index=A_VAL.index[0], col='value')
assert A_VAL == row.value
B_VAL = B.loc[(B.year == row.year) & (B.iiso3c == row.iiso3c) & (B.eiso3c == row.eiso3c) & (B.sitc4 == str(row.sitc4))]
if len(B_VAL) == 0:
print "Data not contained in B"
continue
B_VAL = B_VAL.get_value(index=B_VAL.index[0], col='value')
assert B_VAL == row.value_adj
def TestLevel1(self, verbose=True):
"""
Test Level 1 Data Constructors
"""
for dataset in SITC_DATASET_OPTIONS:
if verbose: print "Testing DATASET Definition: %s" % dataset
for data_type in DATA_TYPE:
if verbose: print "Testing DATA_TYPE: %s" % data_type
#-IF Adjust Hong Kong Data then Add Data to the Tuple-#
if SITC_DATASET_OPTIONS[dataset]['adjust_hk'] == True:
SITC_DATASET_OPTIONS[dataset]['adjust_hk'] = (True, self.hkchina_rawdata)
else:
SITC_DATASET_OPTIONS[dataset]['adjust_hk'] = (False, None)
data1 = construct_sitcr2(self.rawdata, data_type=data_type, level=1, **SITC_DATASET_OPTIONS[dataset]) #-Default Options-#
data2 = construct_sitcr2l1(self.rawdata, data_type=data_type, **SITC_DATASET_OPTIONS[dataset])
assert_frame_equal(data1, data2)
def TestLevel1(self, verbose=True):
"""
Test Level 1 Data Constructors
"""
for dataset in SITC_DATASET_OPTIONS:
if verbose: print "Testing DATASET Definition: %s" % dataset
for data_type in DATA_TYPE:
if verbose: print "Testing DATA_TYPE: %s" % data_type
#-IF Adjust Hong Kong Data then Add Data to the Tuple-#
if SITC_DATASET_OPTIONS[dataset]['adjust_hk'] == True:
SITC_DATASET_OPTIONS[dataset]['adjust_hk'] = (
True, self.hkchina_rawdata)
else:
SITC_DATASET_OPTIONS[dataset]['adjust_hk'] = (False, None)
data1 = construct_sitcr2(
self.rawdata,
data_type=data_type,
level=1,
**SITC_DATASET_OPTIONS[dataset]) #-Default Options-#
data2 = construct_sitcr2l1(self.rawdata,
data_type=data_type,
**SITC_DATASET_OPTIONS[dataset])
assert_frame_equal(data1, data2)
DATA_OPTIONS[dataset]['adjust_hk'] = (
True, rawdata_hk
) #Add Data for construct_sitcr2l3 function, then remove so not saved as an attribute
else:
DATA_OPTIONS[dataset]['adjust_hk'] = (False, None)
#-Intertemporal Product Codes-#
if DATA_OPTIONS[dataset]['intertemp_productcode']:
DATA_OPTIONS[dataset]['intertemp_productcode'] = (True,
ICP[4])
else:
DATA_OPTIONS[dataset]['intertemp_productcode'] = (False,
None)
#-Compute Data-#
data = construct_sitcr2(rawdata.copy(deep=True),
data_type=data_type,
level=4,
values_only=True,
**DATA_OPTIONS[dataset])
store.put(dataset, data, format='table')
#-Remove Hong Kong Data-#
if DATA_OPTIONS[dataset]['adjust_hk'][0] == True:
DATA_OPTIONS[dataset]['adjust_hk'] = True
else:
DATA_OPTIONS[dataset]['adjust_hk'] = False
if DATA_OPTIONS[dataset]['intertemp_productcode'][0] == True:
DATA_OPTIONS[dataset]['intertemp_productcode'] = True
else:
DATA_OPTIONS[dataset]['intertemp_productcode'] = False
store.get_storer(dataset).attrs.options = DATA_OPTIONS[dataset]
store.get_storer(dataset).attrs.data_type = data_type
store.get_storer(
store = pd.HDFStore(target_dir+fn, complevel=9, complib='zlib')
#-Compute Datasets-#
for dataset in sorted(DATA_OPTIONS.keys()):
print "\n[SITCR2L4] Computing Dataset %s for %s" % (dataset, data_type)
#-IF Adjust Hong Kong Data then Add Data to the Tuple-#
if DATA_OPTIONS[dataset]['adjust_hk'] == True:
DATA_OPTIONS[dataset]['adjust_hk'] = (True, rawdata_hk) #Add Data for construct_sitcr2l3 function, then remove so not saved as an attribute
else:
DATA_OPTIONS[dataset]['adjust_hk'] = (False, None)
#-Intertemporal Product Codes-#
if DATA_OPTIONS[dataset]['intertemp_productcode']:
DATA_OPTIONS[dataset]['intertemp_productcode'] = (True, ICP[4])
else:
DATA_OPTIONS[dataset]['intertemp_productcode'] = (False, None)
#-Compute Data-#
data = construct_sitcr2(rawdata.copy(deep=True), data_type=data_type, level=4, values_only=True, **DATA_OPTIONS[dataset])
store.put(dataset, data, format='table')
#-Remove Hong Kong Data-#
if DATA_OPTIONS[dataset]['adjust_hk'][0] == True:
DATA_OPTIONS[dataset]['adjust_hk'] = True
else:
DATA_OPTIONS[dataset]['adjust_hk'] = False
if DATA_OPTIONS[dataset]['intertemp_productcode'][0] == True:
DATA_OPTIONS[dataset]['intertemp_productcode'] = True
else:
DATA_OPTIONS[dataset]['intertemp_productcode'] = False
store.get_storer(dataset).attrs.options = DATA_OPTIONS[dataset]
store.get_storer(dataset).attrs.data_type = data_type
store.get_storer(dataset).attrs.description = DATA_DESCRIPTION[dataset]
print store
#-Close-#
