Abstract
This paper derives a new method for comparing the weak-form efficiency of markets. The author derives the formula of the Sharpe ratio from the ARMA-GARCH model and finds that the Sharpe ratio just depends on the coefficients of the AR and MA terms and is not affected by the GARCH process. For empirical purposes, the Sharpe ratio can be formulated with a monotonic increasing function of R-squared if the sample size is large enough. One can utilize the Sharpe ratio to compare weak-form efficiency among different markets. The results of stochastic simulation demonstrate the validity of the proposed method. The author also constructs empirical AR-GARCH models and computes the Sharpe ratio for S&P 500 Index and the SSE Composite Index.
Introduction
The Efficient Market Hypothesis (EMH) is often discussed as a binary state—a market is either efficient or it is not. However, in financial reality, efficiency exists on a spectrum. While traditional statistical tests like autocorrelation or variance ratios provide insights into price dependencies, they often fail to capture the economic essence of the problem: the potential for risk-adjusted profit. This paper introduces a novel framework for measuring relative market efficiency by deriving the Sharpe ratio directly from ARMA-GARCH time-series models.
Methodology
The authors bridge the gap between econometrics and finance by deriving closed-form expressions for the expected excess return and its standard deviation based on a simple directional trading strategy (buying when the forecast is positive and selling when negative).
The Sharpe Ratio and GARCH Neutrality
A key theoretical insight is that the Sharpe ratio (SR) depends solely on the AR and MA coefficients of the return process. Interestingly, while a GARCH process affects the variance of the error term (volatility clustering), it does not change the theoretical Sharpe ratio of the forecast strategy. This implies that risk and return scale proportionally, making the Sharpe ratio a stable metric for comparing markets regardless of their specific volatility dynamics.
The Link to Predictability ($R^2$)
For large samples, the authors demonstrate that the Sharpe ratio is a monotonic increasing function of the model’s R-squared ($R^2$). Specifically, the relationship is approximated by the formula $SR \approx \sqrt{2R^2 / (\pi - 2R^2)}$. This provides a direct, quantifiable link between a model's explanatory power and its trading potential.
Empirical Results
The paper validates these theoretical findings through both synthetic simulations and empirical data from two distinct indices (2000–2014): the S&P 500 (representing developed markets) and the SSE Composite Index (representing emerging markets).
- S&P 500: Potential Sharpe ratio of 0.055.
- SSE Composite: Potential Sharpe ratio of 0.086.
Consequently, the S&P 500 is deemed more efficient. Furthermore, the study observed that market predictability—and thus inefficiency—tends to increase during periods of extreme economic stress, such as the 2001 tech bubble burst and the 2008 financial crisis.
Conclusion and Limitations
By using the Sharpe ratio as a yardstick, researchers can rank markets by their degree of weak-form efficiency. While this method assumes a frictionless environment (ignoring transaction costs), it offers a more robust and scale-independent comparison than traditional Q-statistics. Future research may refine this by incorporating trading costs and exploring non-linear models.
Summary updated