"""Table III col 5: does the paper's log likelihood (-233) come from the 573-obs sample?

Run from the package root:  python3 replication/check_t3col5.py data/movie.dta
"""
import sys
import numpy as np
import pandas as pd
import statsmodels.api as sm

d = pd.read_stata(sys.argv[1], convert_categoricals=False)
d = d[(d.pre != 1) & (d.month != 12)]
d["weekend"] = d.opendate + 6 - d.openwday
X = [f"year{i}" for i in range(2, 10)] + ["fourday"] + [f"month{i}" for i in range(1, 11)]
for label, s in (("575 obs", d), ("573 obs (drop documentary, noir)", d[(d.document != 1) & (d.noir != 1)])):
    r = sm.Probit(s.dur.astype(float), sm.add_constant(s[X].astype(float))).fit(disp=0)
    print(label, "N", int(r.nobs), "llf", round(r.llf, 1), "pseudoR2", round(r.prsquared, 3), "null llf", round(r.llnull, 1))