Forecasting Dividend Rebound Probability with Ex-Dividend Event Studies

Preface

A stock is said to “rebound after dividends” when its price climbs back to or above the closing price before the ex-dividend date, meaning that investors not only receive cash dividends but also avoid a capital loss from the price adjustment. This makes dividend-paying stocks with high rebound potential attractive to investors seeking both income and capital preservation.

This study applies the event study methodology to test whether a company’s historical ex-dividend rebound probability can signal future rebound potential, and whether this probability can serve as a basis for selecting high-yield stocks.

TEJ Financial database-A Comprehensive Source for Dividend Events in Taiwan

The data used in this research is sourced from the TEJ Financial Database, which offers complete and structured data on listed companies’ dividend policies and ex-dividend events. Key datasets include:

  • Shareholders’ Meeting Information
    Tracks key dates surrounding shareholder meetings, such as book closure periods and margin trading cutoffs, helping investors align with important corporate actions.
  • Dividend Policy Data
    Details each company’s dividend distribution method (cash or stock), payout amounts, ex-dividend dates, and actual payment dates — enabling evaluation of payout consistency and financial health.
  • Ex-Dividend Performance Data
    Provides stock returns five days before and after ex-dates, time required to fully rebound, and the market index impact, making it easy to analyze rebound speed and investor sentiment.

Backtesting with TEJ Ex-Dividend Data

  • Stock Pool: All listed and OTC common stocks (including KY stocks) that issued dividends.
  • Period: January 1, 2013 – May 1, 2025.
  • Sources: TEJ Ex-Dividend Data, Financial Statements, Company Fundamentals, and Stock Price Data.

Defining the Rebound Probability Factor

To evaluate whether past rebound success predicts future performance:

  • Rebound Probability (cover_prob) is calculated as the number of successful post-dividend rebounds divided by the total dividend events per company.
  • Lagged Rebound Probability (last_cover_prob) represents each stock’s most recent rebound success rate and serves as the grouping factor.

After excluding stocks lacking recent rebound records, 13,640 valid entries remained.

Using Event Study to Measure Abnormal Returns

The event study framework is a standard approach for analyzing how stock prices react to specific corporate events. It does so by calculating abnormal returns(AR), defined as the difference between a stock’s return and that of a market index. When these differences are aggregated over a period, they form the cumulative abnormal return (CAR), which captures the market’s overall reaction.
This study uses the Market Index Adjustment Model to assess whether stocks outperform the market around the final purchase date (Day T)—the last day investors can buy shares and still receive dividends. By comparing returns before and after this date, we evaluate whether historical rebound probabilities are linked to abnormal price movements.

Event Study Settings

This study examines whether historical dividend rebound probability can guide stock selection—helping investors capture dividends while avoiding post-dividend capital losses. We define the event day (T) as the last day to purchase shares and still receive dividends.

To measure market reaction, we use a 21-day event window before and after T (T–21 to T+21). For each stock, we calculate:

  • bef_ret: cumulative abnormal return from T–21 to T–1, and
  • aft_ret: cumulative abnormal return from T to T+21.

To analyze performance differences, stocks are grouped into five brackets based on their most recent rebound probability: 0–20%, 20–40%, 40–60%, 60–80%, and 80–100%. This allows us to compare how stocks with varying rebound histories behave around the ex-dividend period.

Result Analysis

The chat shows stocks with 40–100% rebound probability exhibited significant abnormal returns between T–10 and T–3, with CARs between 0.433% and 0.513%, all statistically significant at the 1% level. This suggests that investors start positioning up to two weeks before T.

Some of this effect may be driven by dividend announcement dates, which often fall two weeks before ex-dividend day.During T–2 to T+2, CARs flattened or declined, indicating that early investors may have exited ahead of retail investors chasing dividends.

After T+3, all high-rebound groups experienced steady selling pressure. CARs turned significantly negative, indicating profit-taking behavior after collecting dividends.

Chart 1, AAR and CAAR around the dividend announcement date ±21 days

Summary Table: CAR (%) Across Time Windows

Rebound Probability (%)CAAR(%) at
T-21~T-11
CAAR(%) at
T-10~T-3
CAAR(%) at
T-2~T2
CAAR(%) at
T3~T21
CAAR(%) at
T-21~T21
0–20-0.085-0.3962.106***-1.2240.401
20–40-0.327-0.2590.239-1.570***-1.918**
40–60-0.1520.462***0.100-0.840***-0.430
60–800.1350.433***0.093-0.583***0.078
80–1000.0180.513***0.170***-0.446***0.254*
(* p < 10%; ** p < 5%; *** p < 1%)

Deepening the Analysis: Rebound Probability × Operating Profit Growth

To uncover more precise market signals, this study cross-references dividend rebound probability with operating income growth, a fundamental indicator of a company’s core business strength.

Results show that among stocks with a rebound probability above 40%, those with positive operating income growth attract earlier and more persistent buying behavior. Their cumulative abnormal returns (CARs) are significantly stronger in the T–10 to T–3 period and show less downside in the post-dividend T+3 to T+21 period. In contrast, stocks with negative growth, even if historically strong at rebounding, experience a steeper selloff after T+3—suggesting that fundamental weakness undermines investor confidence.

Moreover, the T+3 to T+21 CARs for negative-growth stocks are all significantly negative (ranging from -0.79% to -1.19%), highlighting the market’s skepticism about their ability to sustain performance post-dividend. The data also shows that price differentiation between positive and negative growth stocks becomes more pronounced in the pre-dividend period, especially for those with higher rebound probabilities.

Summary Table: CAAR (%) by Rebound Probability × Operating Income Growth

(Only selected groups shown for clarity)

Rebound Probability (%)Operating Income GrowthT-21~T-11 T-10~T-3 T-2~T+2 T+3~T+21 T-21~T+21 
0-20 %Negative0.798 1.824 2.144* -5.268* -0.501 
Positive-0.615 -1.728 2.084* 1.202 0.942 
20-40 %Negative-1.095*** -0.562 -0.020 -2.380** -4.058*** 
Positive0.539 0.084 0.531* -0.654 0.500 
40-60 %Negative-0.399* 0.235 0.171 -1.194*** -1.187*** 
Positive0.137 0.730*** 0.016 -0.424 0.459 
60-80 %Negative-0.248* 0.053 0.097 -0.907*** -1.004*** 
Positive0.522*** 0.816*** 0.090 -0.256 1.172*** 
80-100Negative-0.407*** 0.167* 0.080 -0.785*** -0.945*** 
Positive0.372*** 0.800*** 0.245*** -0.166 1.251*** 
*** p < 0.01; ** p < 0.05; * p < 0.10

Conclusions

Overall, stocks with a higher historical dividend rebound probability (40% to 100%) tend to experience a significant surge in buying activity in the one to two weeks prior to the final purchase date (T–10 to T–3), driving cumulative abnormal returns above the market average. When these signals are further filtered by operating income growth, the strength of the pre-event performance becomes even more pronounced.

However, in the post-dividend period (T+3 to T+21), even stocks with high rebound probability may underperform the market if the company lacks fundamental improvement. This suggests that dividend strategies relying solely on historical rebound trends may be vulnerable to reversal if not supported by strong core operations.

Therefore, historical rebound probability serves as a valuable screening tool, but it is the combination of past market behavior and current financial strength that offers the most reliable foundation for identifying dividend stocks with lasting return potential.

From Research to Practice: How TEJ Enables Event-Driven Strategy Development

This study highlights how combining historical rebound probability with event study techniques and fundamental filters can enhance alpha generation in Taiwan’s equity market. Success in such analysis depends on access to structured, point-in-time, and programmable data.
TEJ’s Quantitative Solutions offer:

  • Fully annotated Ex-Dividend Event Data for tracking rebound outcomes.
  • Point-in-time dividend policy and financial indicators for clean factor modeling.
  • High-performance APIs to support automated screening and backtesting.

Whether you’re building dividend-based alpha models or event-driven trading systems, TEJ’s rich and reliable datasets provide the foundation to move from insight to execution.

“Taiwan stock market data, TEJ collect it all.”

Taiwan Economical Journal (TEJ), a financial database established in Taiwan for over 30 years, serves local financial institutions and academic institutions, and has long-term cooperation with internationally renowned data providers, providing high-quality financial data for five financial markets in Asia. 

  • Complete Coverage: Includes all listed companies on stock markets in Taiwan, China, Hong Kong, Japan, Korea, etc. 
  • Comprehensive Analysis of Enterprises: Operational aspects, financial aspects, securities market performance, ESG sustainability, etc. 
  • High-Quality Database: TEJ data is cleaned, checked, enhanced, and integrated to ensure it meets the information needs of financial and market analysis. 

With TEJ’s assistance, you can access relevant information about major stock markets in Asia, such as securities market, financials data, enterprise operations, board of directors, sustainability data, etc., providing investors with timely and high-quality content. Additionally, TEJ offers advisory services to help solve problems in theoretical practice and financial management!

Further Reading:

Back
Processing...