File size: 3,496 Bytes
05707f0 | 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 | import zipfile
import pandas as pd
import pytest
from emft.data import load_jkp, normalise
SITE = pd.DataFrame(
{
"location": ["bra", "bra", "usa", "usa"],
"name": ["be_me", "be_me", "be_me", "be_me"],
"freq": "monthly",
"weighting": "vw_cap",
"direction": 1,
"n_stocks": [100, 100, 3000, 3000],
"n_stocks_min": [30, 3, 900, 900],
"date": ["2020-01-31", "2020-02-29", "2020-01-31", "2020-02-29"],
"ret": [0.01, -0.02, 0.005, 0.0],
}
)
def test_site_layout_is_normalised_and_thin_cells_dropped():
df = normalise(SITE.copy(), min_stocks=5)
assert list(df.columns) == ["country", "factor", "month", "ret", "region"]
assert len(df) == 3 # the bra Feb row has n_stocks_min = 3
assert set(df["region"]) == {"EM", "DM"}
assert str(df["month"].iloc[0]) == "2020-01"
def test_internal_layout_uses_requested_weighting():
raw = pd.DataFrame(
{"excntry": ["CHN"], "characteristic": ["ret_12_1"], "date": ["2021-06-30"],
"ret_ew": [0.03], "ret_vw": [0.02], "ret_vw_cap": [0.01]}
)
assert normalise(raw, weighting="vw").loc[0, "ret"] == pytest.approx(0.02)
assert normalise(raw).loc[0, "country"] == "chn"
def test_percent_returns_are_rejected():
bad = SITE.copy()
bad["ret"] = [1.5, -2.0, 0.8, 0.9]
with pytest.raises(ValueError, match="percent"):
normalise(bad, min_stocks=0)
def test_duplicates_are_rejected():
dup = pd.concat([SITE, SITE.iloc[[0]]])
with pytest.raises(ValueError, match="duplicate"):
normalise(dup, min_stocks=0)
def test_macro_state_variables(tmp_path):
import numpy as np
from emft.data import load_macro
days = pd.date_range("2004-01-01", "2007-12-31", freq="D")
pd.DataFrame({"observation_date": days, "VIXCLS": np.arange(len(days)) % 40 + 10.0}).to_csv(
tmp_path / "VIXCLS.csv", index=False)
pd.DataFrame({"observation_date": days, "T10Y3M": "."}).assign(T10Y3M=lambda d: d["T10Y3M"].where(
d["observation_date"].dt.day != 15, "1.5")).to_csv(tmp_path / "T10Y3M.csv", index=False)
months = pd.date_range("2004-01-01", "2007-12-01", freq="MS") # overlaps the new index, as on FRED
pd.DataFrame({"observation_date": months, "TWEXBMTH": 100 * 1.01 ** np.arange(len(months))}).to_csv(
tmp_path / "TWEXBMTH.csv", index=False)
new_days = pd.date_range("2006-01-01", "2007-12-31", freq="D")
# the new index has a different base but keeps growing 1% a month
level = 50 * 1.01 ** ((new_days.year - 2006) * 12 + new_days.month - 1)
pd.DataFrame({"observation_date": new_days, "DTWEXBGS": level}).to_csv(tmp_path / "DTWEXBGS.csv", index=False)
m = load_macro(tmp_path)
assert m.loc[pd.Period("2004-02", freq="M"), "mac_vix"] == pytest.approx(np.log(10 + (59 % 40))) # Feb 29
assert m.loc[pd.Period("2006-06", freq="M"), "mac_term"] == pytest.approx(1.5) # "." is missing, not zero
# 12-month log change is 12 * log(1.01) on both sides of the 2006-01 splice
assert m.loc["2005-01":"2007-12", "mac_usd12"].to_numpy() == pytest.approx(12 * np.log(1.01))
def test_directory_with_site_zip_is_read(tmp_path):
# jkpfactors.com serves one zip per selection; dropping it in data/raw/ must work
with zipfile.ZipFile(tmp_path / "[all_countries]_[all_factors]_[monthly]_[vw_cap].zip", "w") as z:
z.writestr("factors.csv", SITE.to_csv(index=False))
df = load_jkp(tmp_path, min_stocks=5)
assert len(df) == 3
|