Stock Daily Returns Distribution

import matplotlib.pyplot as plt
import numpy as np
from scipy import stats

np.random.seed(49)

# Stock returns are typically slightly fat-tailed
returns = np.random.standard_t(df=5, size=1000) * 2
returns = np.clip(returns, -15, 15)

kde = stats.gaussian_kde(returns)
x = np.linspace(-15, 15, 500)
y = kde(x)

fig, ax = plt.subplots(figsize=(12, 6), facecolor='#0a0a0f')
ax.set_facecolor('#0a0a0f')

# Color positive vs negative
pos_mask = x >= 0
neg_mask = x < 0

ax.fill_between(x[neg_mask], y[neg_mask], alpha=0.5, color='#F5276C')  # Coral for losses
ax.fill_between(x[pos_mask], y[pos_mask], alpha=0.5, color='#6CF527')  # Lime for gains
ax.plot(x, y, color='white', linewidth=2.5)

# Add normal distribution overlay for comparison
normal_y = stats.norm.pdf(x, 0, np.std(returns))
ax.plot(x, normal_y, color='#F5B027', linewidth=2, linestyle='--', alpha=0.7, label='Normal Dist.')

# Key statistics
ax.axvline(0, color='#e2e8f0', linestyle='-', linewidth=1, alpha=0.5)
ax.axvline(np.mean(returns), color='#27D3F5', linestyle='--', linewidth=2, label=f'Mean: {np.mean(returns):.2f}%')

# VaR lines
var_95 = np.percentile(returns, 5)
ax.axvline(var_95, color='#F5276C', linestyle=':', linewidth=2, label=f'VaR 95%: {var_95:.1f}%')

ax.set_xlabel('Daily Return (%)', fontsize=12, color='white', fontweight='500')
ax.set_ylabel('Density', fontsize=12, color='white', fontweight='500')
ax.set_title('Stock Market Daily Returns Distribution', fontsize=16, color='white', fontweight='bold', pad=15)

ax.tick_params(colors='white', labelsize=10)
for spine in ax.spines.values():
    spine.set_color('#334155')
ax.legend(loc='upper right', facecolor='#1e293b', edgecolor='#334155', labelcolor='white')
ax.grid(True, alpha=0.1, color='white')

plt.tight_layout()
plt.show()