Scatterplot

# === IMPORTS ===
import pandas as pd
import numpy as np
import matplotlib.pyplot as plt
import seaborn as sns
from scipy import stats

# === USER-EDITABLE PARAMETERS ===
# Change: Column names for the plot
x_column = 'Height'  # Change: X-axis variable name
y_column = 'Weight'  # Change: Y-axis variable name
hue_column = 'Gender'  # Change: Color grouping variable

# Change: Plot appearance
figsize = (10, 8)  # Change: Figure size as (width, height)
title_fontsize = 18  # Change: Title font size
label_fontsize = 16  # Change: Axis label and tick font size
male_color = '#1f77b4'  # Change: Color for male points (hex code)
female_color = '#ff7f0e'  # Change: Color for female points (hex code)
male_marker = 'o'  # Change: Marker shape for male points
female_marker = 's'  # Change: Marker shape for female points
alpha = 0.7  # Change: Transparency of points (0-1)
show_regression = True  # Change: Set to False to hide regression lines

# Change: Dataset parameters
n_samples = 200  # Change: Number of data points to generate
random_seed = 42  # Change: Random seed for reproducibility

# === DATA GENERATION ===
# Set random seed for reproducibility
np.random.seed(random_seed)

# Generate synthetic dataset
n_male = n_samples // 2
n_female = n_samples - n_male

# Male data: taller and heavier on average
male_height = np.random.normal(178, 8, n_male)
male_weight = 0.8 * male_height + np.random.normal(10, 5, n_male)

# Female data: shorter and lighter on average
female_height = np.random.normal(165, 7, n_female)
female_weight = 0.7 * female_height + np.random.normal(5, 4, n_female)

# Combine into DataFrame
df = pd.DataFrame({
    'Height': np.concatenate([male_height, female_height]),
    'Weight': np.concatenate([male_weight, female_weight]),
    'Gender': ['Male'] * n_male + ['Female'] * n_female
})

# === STATISTICAL ANALYSIS ===
# Initialize variables
male_slope, male_intercept, male_r, male_p, male_std_err = None, None, None, None, None
female_slope, female_intercept, female_r, female_p, female_std_err = None, None, None, None, None

# Calculate regression statistics for each gender
try:
    male_data = df[df['Gender'] == 'Male']
    female_data = df[df['Gender'] == 'Female']
    
    male_slope, male_intercept, male_r, male_p, male_std_err = stats.linregress(
        male_data[x_column], male_data[y_column]
    )
    
    female_slope, female_intercept, female_r, female_p, female_std_err = stats.linregress(
        female_data[x_column], female_data[y_column]
    )
    
    # Print statistical results
    print("\n=== REGRESSION ANALYSIS ===")
    print(f"Male regression: Weight = {male_slope:.2f} * Height + {male_intercept:.1f}")
    print(f"  R² = {male_r**2:.3f}, p-value = {male_p:.3e}")
    print(f"Female regression: Weight = {female_slope:.2f} * Height + {female_intercept:.1f}")
    print(f"  R² = {female_r**2:.3f}, p-value = {female_p:.3e}")
    print(f"\nDataset summary: {len(df)} total points ({n_male} male, {n_female} female)")
    
except Exception as e:
    print(f"Note: Statistical analysis could not be completed - {str(e)}")

# === PLOT CREATION ===
# Create figure with proper margins
fig, ax = plt.subplots(figsize=figsize)
plt.subplots_adjust(top=0.92, bottom=0.12, left=0.12, right=0.95)

# Plot scatter points with redundant encoding (color + shape)
male_data = df[df['Gender'] == 'Male']
female_data = df[df['Gender'] == 'Female']

ax.scatter(male_data[x_column], male_data[y_column], 
           c=male_color, marker=male_marker, alpha=alpha, 
           s=60, label='Male', edgecolors='white', linewidth=0.5)

ax.scatter(female_data[x_column], female_data[y_column], 
           c=female_color, marker=female_marker, alpha=alpha, 
           s=60, label='Female', edgecolors='white', linewidth=0.5)

# Add regression lines if enabled
if show_regression and male_slope is not None and female_slope is not None:
    # Generate x values for regression lines
    x_range = np.linspace(df[x_column].min(), df[x_column].max(), 100)
    
    # Male regression line
    male_y_pred = male_slope * x_range + male_intercept
    ax.plot(x_range, male_y_pred, color=male_color, linestyle='--', 
            linewidth=2, alpha=0.8, label=f'Male trend (R²={male_r**2:.2f})')
    
    # Female regression line
    female_y_pred = female_slope * x_range + female_intercept
    ax.plot(x_range, female_y_pred, color=female_color, linestyle='--', 
            linewidth=2, alpha=0.8, label=f'Female trend (R²={female_r**2:.2f})')

# === PLOT STYLING ===
# Set labels and title
ax.set_xlabel('Height (cm)', fontsize=label_fontsize)
ax.set_ylabel('Weight (kg)', fontsize=label_fontsize)
ax.set_title('Weight increases with height for both genders, males show stronger correlation', 
             fontsize=title_fontsize, pad=20)

# Set tick label sizes
ax.tick_params(labelsize=label_fontsize)

# Add grid for better readability
ax.grid(True, alpha=0.3, linestyle='-', linewidth=0.5)

# Add legend with proper positioning
legend = ax.legend(loc='upper left', fontsize=label_fontsize-2, 
                   framealpha=0.9, edgecolor='gray', fancybox=True)



# Set axis limits with some padding
x_min, x_max = df[x_column].min(), df[x_column].max()
y_min, y_max = df[y_column].min(), df[y_column].max()
ax.set_xlim(x_min - 5, x_max + 5)
ax.set_ylim(y_min - 5, y_max + 5)

# Apply tight layout
plt.tight_layout()

# Display the plot
plt.show()
# END-OF-CODE

Correlation: 0.891
R-squared: 0.794
P-value: < 0.001
Regression equation: Weight = 0.599*Height - 50.234