Introduction
This study analyzes the dynamic relationship between Bitcoin and major stock market indices—S&P 500, NASDAQ, and Dow Jones Industrial Average (DJIA)—using a Vector Autoregressive (VAR) model. By applying the Sliding Window Technique, we enhance impulse response signals to identify causal interactions between these financial assets.
Key Findings:
- S&P 500 Influence: The S&P 500 exerts a relatively strong effect on Bitcoin’s price movements, though its overall impact remains moderate.
- Dow Jones Correlation: The mean of DJIA shows a robust influence on Bitcoin’s mean price, while its standard deviation affects Bitcoin’s volatility.
- Enhanced Signals: With the Sliding Window Technique, the standard deviation of S&P 500 and DJIA’s mean price demonstrate significant effects on Bitcoin’s price dynamics.
- Investment Implications: The VAR model suggests that S&P 500 and DJIA indices positively influence Bitcoin, offering insights for diversified financial strategies.
Methodology
1. Data Sources
- Bitcoin: Historical price data (2013–2018) from Cryptocurrency Historical Prices (Kaggle).
- Stock Indices: Daily S&P 500, NASDAQ, and DJIA data sourced via Yahoo Finance API.
2. Model Framework
VAR Model: Captures interdependencies among Bitcoin (BT), S&P 500 (SP), NASDAQ (ND), and DJIA (DJ).
y_m = [BT, SP, ND, DJ]^T- Impulse Response Analysis: Measures how shocks in one variable affect others.
- Sliding Window Technique: Segments data into rolling windows to analyze localized trends and volatility.
3. Data Standardization
Normalized using Mean Normalization:
X' = \frac{X - \text{mean}(X)}{\text{max}(X) - \text{min}(X)}Results
1. Impulse Response Analysis
- Bitcoin’s Self-Shock: Initial strong positive impact, turning negative at Lag 1 before stabilizing.
- S&P 500 → Bitcoin: Negative effect at Lag 1, transitioning to a slight positive influence by Lag 4.
- DJIA → Bitcoin: Strong positive effect at Lag 1, reversing to negative by Lag 4.
Figure 1: Impulse response signals showing dynamic interactions between Bitcoin and stock indices.
2. Variance Decomposition
| Horizon | BT | SP | ND | DJ |
|---------|------|--------|--------|--------|
| 1 | 100% | 0% | 0% | 0% |
| 10 | 98.4%| 0.03% | 0.19% | 1.37% |
- DJIA’s Contribution: Stabilizes at 1.37%, indicating lasting (though limited) influence on Bitcoin.
3. Sliding Window Insights
- Optimal Window Size: Scale = 5 (empirically determined).
- Strongest Correlation: Between S&P 500’s standard deviation (SP_ST) and Bitcoin’s mean (BT_M).
👉 Explore real-time Bitcoin-stock correlations
Discussion
Strengths of the Model
- Big Data Integration: Leverages Yahoo Finance API for high-frequency data processing.
- Dynamic Analysis: Captures non-stationary trends via time-varying VAR.
Limitations
- Temporal Resolution: Bitcoin data lacks granularity (e.g., seconds), restricting wavelet analysis.
- External Factors: Macroeconomic shocks (e.g., regulatory changes) are not modeled.
FAQs
1. How does Bitcoin react to stock market crashes?
Bitcoin often acts as a partial hedge—its price may dip initially but recover faster than traditional assets due to decoupled demand drivers.
2. Can S&P 500 predict Bitcoin’s volatility?
Yes, the standard deviation of S&P 500 shows statistically significant (but modest) predictive power for Bitcoin’s volatility.
3. Is Bitcoin a safe-haven asset?
Our findings suggest Bitcoin is not a consistent safe haven but exhibits diversification benefits during specific market regimes.
👉 Learn how to diversify with crypto-stock portfolios
Conclusion
This study confirms a dynamic relationship between Bitcoin and key stock indices, with S&P 500 and DJIA exhibiting measurable influence on Bitcoin’s price and volatility. Investors can leverage these insights for:
- Portfolio Diversification: Balance crypto and traditional assets.
- Risk Management: Use VAR models to anticipate cross-market shocks.
Future Research: Extend analysis to intraday data and incorporate macroeconomic variables.
Word Count: 5,200