import warnings; warnings.filterwarnings("ignore")  # Suppress warnings for clean output
import numpy as np
import pandas as pd
np.random.seed(42)

# Data setup (Airline Passenger Flights Dataset)
import seaborn as sns
df = sns.load_dataset('flights')
print(df.shape)
print(df.head(3))

# Combine 'year' and 'month', and set as new datetime index
df['date'] = pd.to_datetime(df['year'].astype(str) + '-' + df['month'].astype(str) + '-01')
df = df.set_index('date')
print(df.head(3))

# Plotting original time series: monthly passengers
import matplotlib.pyplot as plt
df['passengers'].plot(figsize=(10, 4), title='Monthly Airline Passengers')
plt.ylabel('Passengers')
plt.xlabel('Date')
plt.show()

# Downsample: resample monthly data to yearly by summing passengers
annual = df['passengers'].resample('Y').sum()
print(annual.head())

# Plot the annual passenger totals
annual.plot(kind='bar', figsize=(7,4), title='Yearly Total Passengers')
plt.ylabel('Passengers')
plt.xlabel('Year')
plt.tight_layout()
plt.show()

# Upsample: move from monthly to daily (with forward fill for missing data)
daily = df['passengers'].resample('D').ffill()
print(daily[:10])

# Create a new column showing previous month's passengers (shift by 1 row)
df['previous_month'] = df['passengers'].shift(1)
print(df[['passengers', 'previous_month']].head(5))

# Calculate percent change from previous month
df['percent_change'] = df['passengers'].pct_change() * 100
print(df[['passengers', 'percent_change']].head())

# Plot percent monthly change
df['percent_change'].plot(figsize=(10,4), grid=True, title='Percent Change vs. Previous Month')
plt.ylabel('Percent change')
plt.xlabel('Date')
plt.axhline(0, color='grey', linestyle='--')
plt.show()

# Compute rolling 12-month average of passengers
df['rolling_mean_12'] = df['passengers'].rolling(window=12).mean()
print(df[['passengers', 'rolling_mean_12']].tail(13))

# Plot original passengers and rolling mean
plt.figure(figsize=(10,4))
plt.plot(df.index, df['passengers'], label='Monthly Passengers')
plt.plot(df.index, df['rolling_mean_12'], label='12-Month Moving Avg', linewidth=2)
plt.legend()
plt.title('Passengers vs. 12-Month Rolling Average')
plt.ylabel('Passengers')
plt.xlabel('Date')
plt.show()

# Rolling standard deviation (shows how much results vary month to month)
df['rolling_std_6'] = df['passengers'].rolling(window=6).std()
print(df[['passengers', 'rolling_std_6']].tail(8))

# Combine shifting and rolling: month-over-month change in rolling average
df['rolling_mean_diff'] = df['rolling_mean_12'].diff()
print(df[['rolling_mean_12', 'rolling_mean_diff']].tail())

# Find rolling 6-month averages and highlight the period with the highest value
df['rolling_mean_6'] = df['passengers'].rolling(window=6).mean()
max_idx = df['rolling_mean_6'].idxmax()
max_period = df.loc[max_idx- pd.DateOffset(months=5):max_idx]
print(max_period[['passengers', 'rolling_mean_6']])