Paper: arXiv 2310.04125

Authors: Maria Kulikova, Gennady Kulikov

Abstract

This paper explores a time-varying version of weak-form market efficiency that is a key component of the so-called Adaptive Market Hypothesis (AMH). One of the most common methodologies used for modeling and estimating a degree of market efficiency lies in an analysis of the serial autocorrelation in observed return series. Under the AMH, a time-varying market efficiency level is modeled by time-varying autoregressive (AR) process and traditionally estimated by the Kalman filter (KF). Being a linear estimator, the KF is hardly capable to track the hidden nonlinear dynamics that is an essential feature of the models under investigation. The contribution of this paper is threefold. We first provide a brief overview of time-varying AR models and estimation methods utilized for testing a weak-form market efficiency in econometrics literature. Secondly, we propose novel accurate estimation approach for recovering the hidden process of evolving market efficiency level by the extended Kalman filter (EKF). Thirdly, our empirical study concerns an examination of the Standard and Poor’s 500 Composite stock index and the Dow Jones Industrial Average index. Monthly data covers the period from November 1927 to June 2020, which includes the U.S. Great Depression, the 2008-2009 global financial crisis and the first wave of recent COVID-19 recession. The results reveal that the U.S. market was affected during all these periods, but generally remained weak-form efficient since the mid of 1946 as detected by the estimator.

Complexity vs Empirical Score

  • Math Complexity: 7.5/10
  • Empirical Rigor: 6.0/10 (capped from 8.0: no out-of-sample test; see penalty flags)
  • Quadrant: Holy Grail — high math complexity, high empirical rigor (effective empirical rigor capped from 8.0)

Why this score: The paper employs advanced mathematical techniques including time-varying autoregressive models, state-space representations, and extended Kalman filtering with parameter estimation via maximum likelihood. It demonstrates high empirical rigor through extensive backtesting on major indices (S&P 500, DJIA) over nearly a century of data, including stress periods like the Great Depression and 2008 crisis.

Research Flowchart

  flowchart TD
  A["Research Goal<br>Estimate market efficiency<br>via time-varying AR models"] --> B["Data Selection<br>SP500 & DJIA Indices<br>Monthly: Nov 1927 - Jun 2020"]
  
  B --> C["Methodology<br>Extended Kalman Filter EKF<br>vs. Traditional Kalman Filter"]
  
  C --> D{"Computational Process<br>Estimate Time-Varying AR Coefficients"}
  D --> E["Track Hidden Dynamics<br>Nonlinear market efficiency<br>evolution"]
  
  E --> F["Key Findings Outcomes"]
  
  F --> G["U.S. Market Impact<br>Great Depression, 2008-09 Crisis, COVID-19"]
  F --> H["Weak-form Efficiency<br>Established since mid-1946"]
  
  style A fill:#e1f5fe
  style B fill:#f3e5f5
  style C fill:#e8f5e8
  style D fill:#fff3e0
  style E fill:#fce4ec
  style F fill:#e8eaf6
  style G fill:#ffebee
  style H fill:#e0f2f1

Evidence Card

What the authors claim, on what data, and who has checked it. Read from the paper's own text by a language model: every quote is verbatim, nothing was re-run. How to read this card

Proposes

The paper proposes a novel accurate estimation approach for recovering the hidden process of evolving market efficiency level using the extended Kalman filter (EKF).

Source sentence
we propose novel accurate estimation approach for recovering the hidden process of evolving market efficiency level by the extended Kalman filter (EKF).

Assumes 4

  • Return series follows a time‑varying AR model with homoscedastic conditional variance. model
    Source sentence
    A simple approach to design the tests under the AMH is to assume that the return series follows a time‑varying AR model with homoscedastic conditional variance assumption, see, for example, [14].
  • The innovation εₜ is white noise (IID normal with zero mean). statistical
    Source sentence
    εt ∼IID(0, σ2 ε) (1) where yt is the log return series, βi, i = 1, . . ., n are the parameters of the model, which are assumed to be constant over time within the classical EMH concept.
  • Conditional variance follows a stationary GARCH(1,1) process. model
    Source sentence
    where ω > 0, a1 ≥0, b1 ≥0 and a1 + b1 < 1 ensures that the GARCH(1,1) process is stationary.
  • Market efficiency is defined by zero AR coefficients (β₁ = … = βₙ = 0). market
    Source sentence
    Under the EMH, tests for weak‑form market efficiency based on an AR(n) modeling approach yield the following condition: β1 = β2 = . . . = βn = 0.

Data

  • Standard and Poor’s 500 Composite stock index market-prices public
    November 1927 to June 2020 · U.S. stock market
    Source sentence
    our empirical study concerns an examination of the Standard and Poor’s 500 Composite stock index and the Dow Jones Industrial Average index. Monthly data covers the period from November 1927 to June 2020
  • Dow Jones Industrial Average index market-prices public
    November 1927 to June 2020 · U.S. stock market
    Source sentence
    our empirical study concerns an examination of the Standard and Poor’s 500 Composite stock index and the Dow Jones Industrial Average index. Monthly data covers the period from November 1927 to June 2020

Costs

Not applicable The paper reports no trading or execution result.

Finds 3

  • The U.S. market generally remained weak‑form efficient since the mid‑1946, as detected by the EKF estimator. statistical
    Source sentence
    the results reveal that the U.S. market was affected during all these periods, but generally remained weak‑form efficient since the mid of 1946 as detected by the estimator.
  • Two large periods of inefficiency were identified: July 1934‑August 1938 and May 2009‑July 2014. statistical
    Source sentence
    we conclude that the U.S. market is always weak‑form efficient at the 5% significance level, except two large periods of inefficiency detected from July 1934 till August 1938 and from May 2009 till July 2014.
  • The DJIA index recovered to an efficiency regime in September 2010, while the S&P500 remained inefficient until July 2014 at the 5% confidence level. statistical
    Source sentence
    ... the estimated efficiency level of the DJIA index came back to an efficiency regime on September 2010 at the 5% confidence level, meanwhile the S&P500 index remained inefficient until July 2014 at the same 5% confidence level.

Checked by others

No other paper in the archive has been found to support or contradict these findings (compared with 16 related papers). Coverage is partial, so this is neither replication nor refutation.

Code

No repository by the authors found in the paper.

Extracted 2026-11-14 · same card as JSON · A language model chose these items: check the source sentence before relying on one.