Abstract
This study empirically tests the weak form of the Efficient Market Hypothesis (EMH) on Borsa İstanbul (BIST). Weak-form efficiency implies past prices cannot predict future returns, yet the momentum anomaly suggests past winners keep outperforming. Over a ~10-year sample (2016-01-04–2026-07-23), a momentum portfolio formed on past returns delivered roughly 4.2x relative to the BIST 100 index — a pattern inconsistent with strict weak-form efficiency. Results are discussed via behavioral finance and the Adaptive Markets Hypothesis and rest on a historical simulation.
1. Introduction
The EMH (Fama, 1970) holds that prices reflect all available information, making it impossible to systematically beat the market. Yet empirical work documents many ‘anomalies’ implying return predictability. Our question: if a price-based momentum strategy meaningfully beats the index on BIST, how consistent is that with weak-form efficiency?
2. Theoretical Framework and Literature
- Efficiency: Fama (1970, 1991) systematized the forms and tests of EMH.
- Excess volatility: Shiller (1981) showed prices swing far more than fundamentals justify.
- Overreaction: De Bondt & Thaler (1985) found investors overreact, causing mean reversion.
- Momentum: Jegadeesh & Titman (1993) documented past-winner outperformance — key evidence against weak-form efficiency.
- Calendar effects: Rozeff & Kinney (1976) and French (1980).
- Adaptive markets: Lo (2004) — efficiency is dynamic, not static.
3. Data and Methodology
The sample covers liquid BIST equities over ~10 years (2016-01-04–2026-07-23). A momentum signal ranks stocks on past return; the top 20 are held equal-weighted and periodically rebalanced, benchmarked to the BIST 100. Under weak-form efficiency, a purely price-based strategy should not earn risk-adjusted excess returns. Limitations: backtest basis; simplified transaction costs (which can erode momentum profits); single regime.
4. Findings
- Momentum portfolio — Annualized: +141.6% · 5-year: +2652.8% · Sharpe: 2.50 · vs index: 4.2x
- BIST 100 (XU100) — Trailing 12m: 32.9% · ~20x over sample
A price-only momentum portfolio clearly beating the index implies past returns carry information about future returns — i.e., strict weak-form efficiency does not fully hold on BIST, consistent with Jegadeesh & Titman (1993).
5. Discussion
The premium admits a behavioral reading (under-reaction sustains trends) and a risk-based reading (compensation for an unobserved factor). Lo’s (2004) Adaptive Markets Hypothesis reconciles these: efficiency varies with liquidity, competition and regime. A key caveat is transaction costs, which can erode momentum’s edge; nominal BIST returns also reflect inflation, so index-relative interpretation is essential.
6. Conclusion and Limitations
Evidence suggests BIST departs from strict weak-form efficiency and that a momentum premium is observable in this sample. Read cautiously given the backtest basis, simplified costs and single-period coverage. Efficiency is a matter of degree, not all-or-nothing. Not investment advice; past performance does not guarantee future results.
References
- De Bondt & Thaler (1985). Journal of Finance, 40(3).
- Fama, E. F. (1970). Journal of Finance, 25(2).
- Fama, E. F. (1991). Journal of Finance, 46(5).
- French, K. R. (1980). Journal of Financial Economics, 8(1).
- Jegadeesh & Titman (1993). Journal of Finance, 48(1).
- Lo, A. W. (2004). Journal of Portfolio Management, 30(5).
- Rozeff & Kinney (1976). Journal of Financial Economics, 3(4).
- Shiller, R. J. (1981). American Economic Review, 71(3).
