%matplotlib inline

from pathlib import Path
import random

import pandas as pd
import numpy as np

from scipy import stats
import statsmodels.api as sm
import statsmodels.formula.api as smf

import matplotlib.pylab as plt

try:
    import common

    DATA = common.dataDirectory()
except ImportError:
    DATA = Path().resolve() / "data"

WEB_PAGE_DATA_CSV = DATA / "web_page_data.csv"
FOUR_SESSIONS_CSV = DATA / "four_sessions.csv"
CLICK_RATE_CSV = DATA / "click_rates.csv"
IMANISHI_CSV = DATA / "imanishi_data.csv"

session_times = pd.read_csv(WEB_PAGE_DATA_CSV)
session_times.Time = 100 * session_times.Time

ax = session_times.boxplot(by="Page", column="Time", figsize=(4, 4))
ax.set_xlabel("")
ax.set_ylabel("Time (in seconds)")
plt.suptitle("")

plt.tight_layout()
plt.show()

mean_a = session_times[session_times.Page == "Page A"].Time.mean()
mean_b = session_times[session_times.Page == "Page B"].Time.mean()
print(mean_b - mean_a)

# Permutation test example with stickiness
def perm_fun(x, nA, nB):
    n = nA + nB
    idx_B = set(random.sample(range(n), nB))
    idx_A = set(range(n)) - idx_B
    return x.loc[idx_B].mean() - x.loc[idx_A].mean()


nA = session_times[session_times.Page == "Page A"].shape[0]
nB = session_times[session_times.Page == "Page B"].shape[0]
print(perm_fun(session_times.Time, nA, nB))

random.seed(1)
perm_diffs = [perm_fun(session_times.Time, nA, nB) for _ in range(1000)]

fig, ax = plt.subplots(figsize=(5, 5))
ax.hist(perm_diffs, bins=11, rwidth=0.9)
ax.axvline(x=mean_b - mean_a, color="black", lw=2)
ax.text(50, 190, "Observed\ndifference", bbox={"facecolor": "white"})
ax.set_xlabel("Session time differences (in seconds)")
ax.set_ylabel("Frequency")

plt.tight_layout()
plt.show()

print(np.mean(perm_diffs > mean_b - mean_a))

res = stats.ttest_ind(
    session_times[session_times.Page == "Page A"].Time,
    session_times[session_times.Page == "Page B"].Time,
    equal_var=False,
)
print(f"p-value for single sided test: {res.pvalue / 2:.4f}")

tstat, pvalue, df = sm.stats.ttest_ind(
    session_times[session_times.Page == "Page A"].Time,
    session_times[session_times.Page == "Page B"].Time,
    usevar="unequal",
    alternative="smaller",
)
print(f"p-value: {pvalue:.4f}")

random.seed(1)
obs_pct_diff = 100 * (200 / 23739 - 182 / 22588)
print(f"Observed difference: {obs_pct_diff:.4f}%")
conversion = [0] * 45945
conversion.extend([1] * 382)
conversion = pd.Series(conversion)

perm_diffs = [100 * perm_fun(conversion, 23739, 22588) for _ in range(1000)]

fig, ax = plt.subplots(figsize=(5, 5))
ax.hist(perm_diffs, bins=11, rwidth=0.9)
ax.axvline(x=obs_pct_diff, color="black", lw=2)
ax.text(0.06, 200, "Observed\ndifference", bbox={"facecolor": "white"})
ax.set_xlabel("Conversion rate (percent)")
ax.set_ylabel("Frequency")

plt.tight_layout()
plt.show()

print(np.mean([diff > obs_pct_diff for diff in perm_diffs]))

print(pd.read_csv(FOUR_SESSIONS_CSV).head())

four_sessions = pd.read_csv(FOUR_SESSIONS_CSV)

ax = four_sessions.boxplot(by="Page", column="Time", figsize=(4, 4))
ax.set_xlabel("Page")
ax.set_ylabel("Time (in seconds)")
plt.suptitle("")
plt.title("")

plt.tight_layout()
plt.show()

observed_variance = four_sessions.groupby("Page").mean().var()[0]
print("Observed means:", four_sessions.groupby("Page").mean().values.ravel())
print("Variance:", observed_variance)


# Permutation test example with stickiness
def perm_test(df):
    df = df.copy()
    df["Time"] = np.random.permutation(df["Time"].values)
    return df.groupby("Page").mean().var()[0]


print(perm_test(four_sessions))

random.seed(1)
perm_variance = [perm_test(four_sessions) for _ in range(3000)]
print("Pr(Prob)", np.mean([var > observed_variance for var in perm_variance]))

fig, ax = plt.subplots(figsize=(5, 5))
ax.hist(perm_variance, bins=11, rwidth=0.9)
ax.axvline(x=observed_variance, color="black", lw=2)
ax.text(60, 200, "Observed\nvariance", bbox={"facecolor": "white"})
ax.set_xlabel("Variance")
ax.set_ylabel("Frequency")

plt.tight_layout()
plt.show()

model = smf.ols("Time ~ Page", data=four_sessions).fit()

aov_table = sm.stats.anova_lm(model)
print(aov_table)

res = stats.f_oneway(
    four_sessions[four_sessions.Page == "Page 1"].Time,
    four_sessions[four_sessions.Page == "Page 2"].Time,
    four_sessions[four_sessions.Page == "Page 3"].Time,
    four_sessions[four_sessions.Page == "Page 4"].Time,
)
print(f"F-Statistic: {res.statistic / 2:.4f}")
print(f"p-value: {res.pvalue / 2:.4f}")