# Let us turn off warnings for a smoother experience.
import warnings
warnings.filterwarnings("ignore")

# Let us import the libraries we need.
import pandas as pd
import matplotlib.pyplot as plt

# Data setup: let us load monthly airline passenger data.
url = "https://raw.githubusercontent.com/jbrownlee/Datasets/master/airline-passengers.csv"
df = pd.read_csv(url)
print("Shape:", df.shape)
df.head()

# Let us plot the passenger numbers over time.
plt.figure(figsize=(10,4))
plt.plot(df['Passengers'])
plt.title("Airline Passengers Over Time")
plt.xlabel("Month")
plt.ylabel("Number of Passengers")
plt.show()

# Calculate the 3-month rolling mean and add as a new column.
df['RollingMean3'] = df['Passengers'].rolling(window=3).mean()

# Show the result for the first 10 rows
df[['Passengers', 'RollingMean3']].head(10)

# Let us plot the original data and the rolling mean.
plt.figure(figsize=(10,4))
plt.plot(df['Passengers'], label='Original')
plt.plot(df['RollingMean3'], label='3-Month Rolling Mean', color='orange')
plt.title("Passengers and Rolling Mean")
plt.xlabel("Month")
plt.ylabel("Number of Passengers")
plt.legend()
plt.show()

# Calculate the Exponential Moving Average (span=3 months)
df['EMA3'] = df['Passengers'].ewm(span=3, adjust=False).mean()

# Show first 10 values for comparison
df[['Passengers', 'RollingMean3', 'EMA3']].head(10)

# Plot all three: original, rolling mean, E M A.
plt.figure(figsize=(10,4))
plt.plot(df['Passengers'], label='Original')
plt.plot(df['RollingMean3'], label='3-Month Rolling Mean', color='orange')
plt.plot(df['EMA3'], label='EMA (Span=3)', color='green')
plt.title("Rolling Mean vs EMA")
plt.xlabel("Month")
plt.ylabel("Passengers")
plt.legend()
plt.show()

# Try using a bigger window for rolling mean and E M A (12 months).
df['RollingMean12'] = df['Passengers'].rolling(window=12).mean()
df['EMA12'] = df['Passengers'].ewm(span=12, adjust=False).mean()

plt.figure(figsize=(10,4))
plt.plot(df['Passengers'], label='Original', alpha=0.5)
plt.plot(df['RollingMean12'], label='12-Month Rolling Mean', color='red')
plt.plot(df['EMA12'], label='EMA (Span=12)', color='purple')
plt.title("Yearly Rolling Mean and EMA")
plt.xlabel("Month")
plt.ylabel("Passengers")
plt.legend()
plt.show()

# Using input to pick a window size for rolling mean.
window = int(input("Enter a window size (number of months) for rolling mean: "))
df['UserRollingMean'] = df['Passengers'].rolling(window=window).mean()

plt.figure(figsize=(10,4))
plt.plot(df['Passengers'], label='Original')
plt.plot(df['UserRollingMean'], label=f'Rolling Mean (Window={window})', color='orange')
plt.title(f'Rolling Mean with Window {window}')
plt.xlabel("Month")
plt.ylabel("Passengers")
plt.legend()
plt.show()

# Check for missing values after rolling mean.
missing = df['UserRollingMean'].isna().sum()
print(f"Missing values in UserRollingMean: {missing}")

# Fill missing values with the column's first valid value.
df['UserRollingMeanFilled'] = df['UserRollingMean'].fillna(method='bfill')
df[['UserRollingMean', 'UserRollingMeanFilled']].head(window+2)

# Rolling mean with an odd window, such as 5 months.
df['RollingMean5'] = df['Passengers'].rolling(window=5, min_periods=1).mean()
df[['Passengers', 'RollingMean5']].head(7)

# Mini project: Daily minimum temperatures dataset.
temp_url = "https://raw.githubusercontent.com/jbrownlee/Datasets/master/daily-min-temperatures.csv"
temps = pd.read_csv(temp_url)
print("Shape:", temps.shape)
temps.head()

# Plot daily temperatures for one year.
temps['Date'] = pd.to_datetime(temps['Date'])
one_year = temps[temps['Date'].dt.year == 1981]
plt.figure(figsize=(10,4))
plt.plot(one_year['Date'], one_year['Temp'], label='Daily Min Temp')
plt.title("Daily Min Temperature (1981)")
plt.xlabel("Date")
plt.ylabel("Temp (C)")
plt.legend()
plt.show()

# Compute and plot 7-day rolling mean and E M A
one_year['Rolling7'] = one_year['Temp'].rolling(window=7).mean()
one_year['EMA7'] = one_year['Temp'].ewm(span=7, adjust=False).mean()
plt.figure(figsize=(10,4))
plt.plot(one_year['Date'], one_year['Temp'], label='Daily Min Temp', alpha=0.3)
plt.plot(one_year['Date'], one_year['Rolling7'], label='7-Day Rolling Mean', color='orange')
plt.plot(one_year['Date'], one_year['EMA7'], label='EMA (Span=7)', color='green')
plt.title("7-Day Rolling Mean and E M A of Temperatures")
plt.xlabel("Date")
plt.ylabel("Temp (C)")
plt.legend()
plt.show()

# Challenge: Ask the user for a span value for E M A calculation.
span = int(input("Enter a span (number of days) for EMA: "))
one_year['ChallengeEMA'] = one_year['Temp'].ewm(span=span, adjust=False).mean()
plt.figure(figsize=(10,4))
plt.plot(one_year['Date'], one_year['Temp'], label='Daily Min Temp', alpha=0.3)
plt.plot(one_year['Date'], one_year['ChallengeEMA'], label=f'EMA (Span={span})', color='green')
plt.title(f'Challenge: EMA with Span {span}')
plt.xlabel("Date")
plt.ylabel("Temp (C)")
plt.legend()
plt.show()