International Journal of Data Mining & Knowledge Management Process (IJDKP), Vol.15, No.

3, May 2025

BEYOND WINNERS AND LOSERS: MEDIAN SECTOR

ROTATION IN THE JAPANESE EQUITY MARKET
Sreeya Kotrakona, Natasya Liew and Eugene Pinsky

Department of Computer Science, Boston University Metropolitan College,

Boston, MA02215 USA

ABSTRACT
This study evaluates the effectiveness of a median sector rotation strategy within the Nikkei 500 component
sectors, building on prior research that demonstrated superior risk-adjusted returns by selecting mid-
performing assets. Unlike traditional momentum-based investing, which focuses on winners or losers, the
median strategy systematically reallocates capital to sectors with moderate past performance, reducing
volatility while maintaining steady growth. Our findings reveal that quarterly and semi-annual rebalancing
optimize returns in Japan, differing from U.S.-based studies where monthly rebalancing was more effective.
Unlike buy-and-hold investing, the median strategy tends to outperforms total return and drawdown
reduction, making it a viable alternative for public investors. By applying structured sector rotation rather
than passive indexing, investors gain exposure to Japan’s strongest industries while mitigating downside
risk. The results highlight the strategy’s adaptability across markets and suggest broader applications in
global equities, fixed income, and multi-asset portfolios for enhanced portfolio resilience.

KEYWORDS
median sector rotation, Nikkei 500, risk-adjusted returns, portfolio rebalancing, sector investing, asset allocation,
market volatility, public investment strategy.
.
1. INTRODUCTION
Public investors find complex investment strategies unattractive due to their limited resource in
time, research resources, and expertise. As a result, many defaults to passive investing in broad
market indices such as the S&P 500, prioritizing ease and diversification over-optimization.
While index investing provides stability, it also limits the potential for excess returns and fully
exposes investors to market-wide downturns. A simple, rule-based strategy that adapts
systematically to changing conditions could offer a superior alternative—one that balances return
and risk while remaining accessible to the average investor ([1], [2], [3], [4]).

In this context, our paper extends the "median" sector rotation strategy to the Japanese equity
market, specifically using top component sectors within the Nikkei 500. The foundational
strategy was introduced by Yang and Pinsky, who applied it to U.S. sector ETFs and showed that
selecting the middle-performing sectors — rather than the best or worst from the previous month
consistently outperformed the S&P 500 and equal-weighted strategies in terms of risk-adjusted
returns [5]. Follow-up work recent work by Kumar and Pinsky validated the same principle in the
Dow Jones Industrial Average, further demonstrating that avoiding extremes in asset selection
provides stability and performance enhancement [6].

Adding to this growing body of evidence, a recent study by Sharma, Srivastava, and Pinsky
(2025) applied a data-driven strategy variation to the Indian National Stock Exchange [7]. Their

DOI:10.5121/ijdkp.2025.15301 1
 International Journal of Data Mining & Knowledge Management Process (IJDKP), Vol.15, No. 3, May 2025

research employed clustering and re-ranking of sector indices at different time intervals and
concluded that the middle-performing group yielded the best balance of return and risk—
particularly under annual rebalancing ([2], [3], [4]). Together, these studies underscore the cross-
market robustness of the median rotation strategy.

The rationale behind this approach is straightforward: rather than chase recent winners or hold
onto laggards, the strategy systematically selects assets with moderate past performance. These
mid-tier assets are usually less volatile than top performers and more stable than the bottom ones,
providing a smoother return profile. Unlike broad index investing, which remains fully exposed
to macro-level shocks, median-based rotation offers a more adaptive and risk-conscious
alternative — one that does not require forecasting or deep technical expertise as with other
sector rotation methods ([8]–[10]).

In this study, we test the applicability of this strategy to Japanese component sectors, examining
whether its effectiveness holds in a new market context. By extending prior U.S. and Indian
applications to Japan, we assess the universality of this strategy as a practical alternative to
passive investing — one capable of delivering better long-term resilience and risk-adjusted
returns.

The Related Work section follows this introduction and provides a review of existing research on
sector rotation strategies. The Background section provides further context focusing on the
characteristics of the Japanese equity market. We then outline our Methodology and Dataset
details, followed by the Results and Discussion section, where we evaluate the effectiveness of
the median sector rotation strategy in Japan. Finally, in the Conclusion section, we present a
summary of our findings.

2. RELATED WORK
In this section, we review several similar studies that are closely related to our work. Sector
rotation strategies, which involve selecting sectors based on their past performance to guide
future investments, have been studied in different markets. In [5], the median sector rotation
strategy was introduced as a possible alternative to traditional investing strategies. Rather than
focusing on the top or bottom performers, the strategy focuses on the middle-performing sectors.
Their study utilized sector ETFs in the US market, ranking sectors by their past performance and
rebalancing them monthly. This study revealed that the middle-based strategies resulted in better
performance compared to both the S&P 500 and equal-weight benchmarks.

In [6], the median sector rotation strategy was further explored. The study applied this strategy to
the Dow Jones Industrial Average (DJIA) and the results were documented. The study found that
the median sector rotation strategy consistently outperforms other strategies such as the Dogs of
the Dow strategy which focuses on stocks with the highest dividend yields. The results also
showed that the median sector rotation strategy not only provided higher returns but also reduced
drawdowns.

The application of the median sector rotation strategy was further expanded in [7] where the
strategy was applied to the Indian National Stock Exchange (NSE). In their study, they utilized
data mining techniques to rank sectors at different time frequencies. Their paper concluded that
selecting the middle performing sectors resulted in the best balance of return and risk, particularly
with annual rebalancing.

While these studies show the effectiveness of the median sector rotation in markets like the US
and India, there is limited research on its application in Japan. The study aims to fill that gap by
2
 International Journal of Data Mining & Knowledge Management Process (IJDKP), Vol.15, No. 3, May 2025

testing the median sector rotation strategy with Nikkei 500 and evaluating whether the strategy is
just as effective in Japan.

3. BACKGROUND
3.1. Overview of the Japanese Equity Market

The Japanese stock market offers investors exposure to various industries advanced technology,
automotive manufacturing, financial services, consumer goods, and 32 other sectors. The market
has historically exhibited distinct economic cycles and structural inefficiencies that differentiate it
from other major global equity markets. These characteristics have made investing in Japanese
equities an opportunity and a challenge for institutional and retail investors alike [11].

Since Japan’s asset bubble collapse in the early 1990s, the Japanese market has experienced
prolonged periods of stagnation, mixed with phases of strong economic growth and policy-driven
market rallies. The Bank of Japan (BOJ) monetary easing policies and corporate governance
reforms have contributed to recent improvements in capital efficiency and shareholder returns.
Yet, volatility remains a persistent feature of the market. Unlike the steady expansion seen in the
U.S. equity market, Japanese stocks tend to move in response to global economic cycles, changes
in domestic policy, and investor sentiment. These factors attribute to the fluctuations in sector
performance [12].

3.2. Volatility and Sector Dynamics in Japan’s Stock Market

Despite the seemingly stable market, the Japan’s equity market is highly sensitive to external
economic shocks, currency fluctuations, and domestic policy changes. Using studies on market
sentiment, findings suggest that negative news has a disproportionately strong impact on
Japanese equities [13]. This contributed to pronounced price fluctuations, particularly in small-
cap and mid-cap stocks [12], leading investors to seek strategies that mitigate exposure to high-
volatility stocks while still capturing market trends.

One popular approach in the Japanese market is sector-based investing. This strategy allows
investors to invest in industries that align with different phases of the economic cycle. While the
typical broad market index investing (e.g., Nikkei 225, TOPIX) remains widely used, sector-
based strategies have gained traction as they offer greater flexibility in navigating market
inefficiencies [11] with a basic understanding of current sector performance.

3.3. The Nikkei 500 Index and Its Components

The Nikkei 500 comprises 500 companies across multiple market sectors. Unlike the Nikkei 225,
which is based on stock prices that can be heavily influenced by an outlier of high-priced stocks,
the Nikkei 500 offers a more balanced sector-based distribution. This makes the index more
useful as a benchmark for structured investment strategies [11].

Some of the top sectors represented by the Nikkei 500 include:

 Marine Transport: – This sector includes companies engaged in maritime logistics and
cargo shipping, which are critical to Japan’s export-drive economy and global trade
networks
 Insurance: – This sector includes firms that offer life and non-life insurance to individuals
and businesses.
3
 International Journal of Data Mining & Knowledge Management Process (IJDKP), Vol.15, No. 3, May 2025

 Shipbuilding: – This sector includes companies that design and build ships for
commercial and industrial use.

Given the broader coverage of mid-cap and large-cap stocks, the Nikkei 500, is better suited for
sector rotation strategies compared to the more concentrated Nikkei 225 [11]. Sector rotation
withing the Nikkei 500 offers an alternative to passive market investing.

Given the volatility in sector performance, focusing on middle-performing sectors may provide a
better risk-return balance [14]. By rebalancing towards these sectors, investors can potentially
achieve better long-term returns while minimizing downside risk [11].

This study builds on this concept by applying the median sector rotation strategy to the Nikkei
500, focusing on middle-performing sectors as an alternative to broad market index investing and
momentum-driven sector selection. The next sections will detail the methodology and present the
empirical results.

4. METHODOLOGY AND DATASET

4.1. Data Selection and Sector Composition

To evaluate the effectiveness of a sector-based rotation strategy within the Japanese stock market,
we constructed a portfolio using components from the Nikkei 500 Index. Due to the unbalanced
distribution of companies within the index, we limited our selection to the top 9 sectors by market
capitalization. This ensures the sector rotation strategy is applied to a representative, liquid subset
of the market, avoiding biases from underrepresented sectors.

The selected sectors (with daily data on the corresponding tickers from Yahoo Finance) are:

 Marine Transport: ˆNG30.OS

 Precision Instrument: ˆNG19.OS
 Insurance: ˆNG26.OS
 Mining: ˆNG02.OS
 Shipbuilding: ˆNG16.OS
 Petroleum: ˆNG09.OS
 Communications: ˆNG33.OS
 Other Manufacturing: ˆNG20.OS
 Electric Machinery: ˆNG15.OS

These sectors collectively represent the dominant industries within the Japanese equity market
and capture a broad spectrum of economic cycles. The selection approach is consistent with prior
research that has demonstrated the effectiveness of mid-performance rotation strategies across
diversified sector groups [5].

4.1.1. Data Limitation

One of the primary limitations of this study is the limited timeframe of data. The dataset used in
this analysis spans from 2018 to 2024 as these were the only years publicly accessible. While this
timeframe does capture a range of economic cycles and market conditions, it is possible that a
longer time series could have given a more comprehensive look into the performance of the
median sector rotation strategy across different market phases.

4
 International Journal of Data Mining & Knowledge Management Process (IJDKP), Vol.15, No. 3, May 2025

4.2. Portfolio Construction and Rotation Strategy

Following the methodology outlined in the previous studies [6] on medianbased sector rotation,
we implemented a systematic ranking and rebalancing approach:

1. Sector Performance Ranking

 At the end of each rebalance period, we rank the 9 sectors based on their total return over
the previous period [15].
 Sectors are then categorized into Winners (Top 3), Losers (Bottom 3), and Median
(Middle 3) performers.

2. Portfolio Allocation

 Equal-weighted portfolios are constructed for each of the three groups [16].
 The Median Portfolio comprises the three middle-performing sectors, following prior
research showing that mid-tier assets often provide better risk-adjusted returns compared
to extreme performers.
 The Winner Portfolio consists of the top three best-performing sectors.
 The Loser Portfolio consists of the three worst-performing sectors. • A benchmark Buy-
and-Hold Portfolio (B&H), consisting of all 9 sectors equally weighted without rotation,
is used for comparison.

3. Rebalancing Frequencies

 We tested five rebalancing intervals to evaluate their impact on strategy performance

– Weekly
– Monthly
– Quarterly
– Semi-Annual
– Annual
 These variations allow us to assess the sensitivity of the strategy to different time
horizons.

4.3. Performance Metrics and Risk Measures

To evaluate the effectiveness of each portfolio strategy, we measured

1. Final Portfolio Value – The total value of each portfolio at the end of the evaluation
period.
2. Annualized Returns – The compounded annual growth rate (CAGR) of each portfolio
3. Annualized Volatility – The standard deviation of returns, reflecting portfolio risk.
4. Tracking Error – Deviation from the benchmark performance (B&H portfolio).
5. Sharpe Ratio – A risk-adjusted performance measure.
6. Maximum Drawdown (MDD) – The largest peak-to-trough loss during the investment
period using the "Empyrical" Python library (https://pypi.org/project/empyrical/) to
calculate.

5
 International Journal of Data Mining & Knowledge Management Process (IJDKP), Vol.15, No. 3, May 2025

4.4. Evaluation and Interpretation of Results

The results were analysed to determine:

 Whether the median rotation strategy provides superior risk-adjusted returns compared to
the winner- and loser-based rotations.
 The effectiveness of different re-balancing periods in optimizing returns while
minimizing risk.
 The relative performance of sector-based investing versus buy-and-hold index investing
in the Nikkei 500.

By systematically assessing these metrics across different re-balancing frequencies, we provide

an empirical basis for determining the optimal strategy for sector rotation in the Japanese stock
market. The next section presents our findings and their implications for portfolio construction.

5. RESULTS AND FINDINGS

5.1. Portfolio Growth and Performance Across Re-balancing Frequencies

The final portfolio values indicate that the Winner strategy consistently outperforms the other
strategies, particularly at longer re-balancing intervals. The Median strategy, which selects the
mid-performing sectors, exhibits steady growth with lower risk exposure than both the Winner
and Loser portfolios.

 Winner Strategy: Generates the highest final portfolio value, particularly with annual re-
balancing.
 Median Strategy: Demonstrates moderate but stable growth, positioning itself as a
balanced investment strategy.
 Loser Strategy: Under-performs across all re-balancing periods, reinforcing the
persistence of downward momentum in weak-performing sectors.
 Buy-and-Hold (B&H) Strategy: Lags behind active strategies but provides lower
volatility and stability over time.

Figure 1: Rotation strategy values across different re-balancing frequencies

6
 International Journal of Data Mining & Knowledge Management Process (IJDKP), Vol.15, No. 3, May 2025

These findings indicate that shorter re-balancing periods (for weekly and monthly) will lead to
reduced performance across all strategies, likely due to increased transaction costs and excessive
portfolio churn. Quarterly and semi-annual re-balancing can optimize the Median strategy’s risk-
adjusted returns, while annual re-balancing yields the highest total returns for the Winner
strategy.
5.2. Growth and Annualized Returns

Comparing cumulative growth and annualized returns across different strategies and re-balancing
periods:

 The Winner strategy achieves the highest growth over long-term horizons but suffers
from high volatility and drawdowns.
 The Median strategy outperforms the Buy-and-Hold strategy while maintaining lower
risk exposure than the Winner strategy.
 The Loser strategy fails to recover from underperformance, making it an inefficient
investment approach.

5.3. Risk Analysis: Maximum Drawdowns and Volatility

Maximum drawdown (MDD) values highlight the risk associated with each strategy.

 The Loser strategy experiences the largest drawdowns, supporting the hypothesis that
underperforming sectors continue to decline rather than revert.
 The Winner strategy, despite achieving the highest returns, exhibits significant
drawdowns, suggesting that top-performing sectors are prone to high volatility.
 The Median strategy has the lowest drawdowns across most rebalancing periods,
demonstrating its ability to mitigate downside risk while maintaining long-term stability.

Figure 2: Rotation Strategy Max Drawdowns across different rotational time periods

Volatility measures reinforce these findings:

 Winner and Loser portfolios exhibit the highest volatility, confirming that extreme sector
performance leads to higher risk.
 The Median strategy maintains moderate volatility, reinforcing its potential for stability
and risk-adjusted return optimization.

7
 International Journal of Data Mining & Knowledge Management Process (IJDKP), Vol.15, No. 3, May 2025

Figure 3: Rotation Strategy volatility across different rotational periods

5.4. Annualized Volatility and Tracking Errors

Annualized volatility and tracking errors demonstrate that:

 Winner and Loser strategies have the highest tracking errors relative to the benchmark
(B&H), particularly in short-term re-balancing periods
 The Median strategy maintains moderate tracking errors, making it an attractive option
for investors seeking consistency with reduced deviation from index performance.
 Short-term rebalancing increases volatility across all strategies, reinforcing the
inefficiency of high-frequency portfolio adjustments.

5.5. Risk-Adjusted Performance: Sharpe Ratio Analysis

The Sharpe Ratio, a tool that measures risk-adjusted returns, indicates that:

 The Winner strategy has the highest Sharpe Ratio in longer rebalancing periods,
benefiting from sustained sector momentum.
 The Median strategy consistently achieves competitive Sharpe ratios, particularly with
quarterly and semi-annual re-balancing, suggesting it effectively balances return and risk.
 The Loser strategy underperforms in Sharpe Ratios, confirming that weak-performing
sectors tend to remain weak.
 Buy-and-Hold provides stable but lower Sharpe Ratios, reflecting its lower-risk but
lower-return nature.

Figure 4: Comparison of Average Sharpe Ratios

8
 International Journal of Data Mining & Knowledge Management Process (IJDKP), Vol.15, No. 3, May 2025

5.6. Key Findings

The Median strategy provides the best balance between return and risk, outperforming Buy-and-
Hold while maintaining lower volatility and drawdowns.

 The Winner strategy achieves the highest absolute returns but at the cost of increased
risk, making it suitable for aggressive investors [17].
 The Loser strategy consistently underperforms, reinforcing that weak sectors do not
recover quickly [17].
 Quarterly and semi-annual re-balancing optimizes risk-adjusted performance, particularly
for the Median strategy.
 Annual re-balancing is most effective for maximizing returns in the Winner strategy but
exposes investors to greater drawdowns.
 Frequent re-balancing (weekly, monthly) negatively impacts performance across all
strategies, likely due to excessive turnover and transaction costs.

The results suggest that a sector rotation strategy focused on middle-performing sectors provides
a compelling alternative to both passive index investing and momentum-based approaches. The
following section will discuss the broader implications of these findings for portfolio construction
and market application.

6. DISCUSSION AND FUTURE STUDIES

6.1. Comparison with Previous Studies

The results of our study closely align with prior research on median-based sector rotation
strategies, reinforcing the conclusion that investing in midperforming assets yields an optimal
balance between return and risk. This strategy consistently outperforms both winner- and loser-
focused approaches, offering lower volatility and more stable growth across multiple markets.

Building on Yang and Pinsky (2022), which applied the median strategy to U.S. S&P 500 sector
ETFs, our study confirms its cross-market applicability. While the U.S. study favoured monthly
re-balancing, quarterly and semi-annual rebalancing is more effective in Japan, highlighting how
optimal timing may depend on the underlying market structure.

Our findings also extend the results of Kumar, Valath, and Pinsky (2025), who implemented the
strategy within the Dow Jones Industrial Average (DJIA) [6]. Their work showed that median-
slice rebalancing outperformed both buy-and-hold and winner-focused strategies, a similar
pattern in our Nikkei 500 application.

New evidence from Sharma, Srivastava, and Pinsky further validates these results in the Indian
National Stock Exchange, using 15 sector indices and data mining techniques [7] . Their findings
echoed the same pattern: median groups achieved the highest Sharpe ratios and lowest
drawdowns, with annual re-balancing emerging as the most stable configuration. These three
studies demonstrate that median rotation is robust across developed and emerging markets,
individual stocks and sector indices, and multiple geographic and economic contexts.

A key structural difference lies in the sector composition of the Nikkei 500, which is more
imbalanced than the DJIA or S&P 500, requiring us to focus on the top nine component sectors.
Despite this adjustment, the strategy’s effectiveness remains consistent, demonstrating the
adaptability and resilience of the median rotation framework across diverse index compositions.
9
 International Journal of Data Mining & Knowledge Management Process (IJDKP), Vol.15, No. 3, May 2025

6.2. Implications for Investors and Portfolio Management

Our findings prove that the median sector rotation strategy offers a viable alternative to passive
index investing and traditional momentum-based strategies. By systematically selecting mid-
performing sectors, investors can:

1. Reduce downside risk while still capturing positive market trends.

2. Avoid overexposure to extreme volatility, which often presents itself in winner and loser-
focused strategies.
3. Adapt to market fluctuations without requiring complex investing models or predictive
analytics.

The study also highlights the importance of re-balancing frequency in optimizing portfolio
performance. While the monthly re-balancing strategy was optimal in previous studies using U.S.
sector ETFs, we find that quarterly and semi-annual re-balancing strategies produce better
outcomes in the Nikkei 500 component sector rotation. This suggests that investors should tailor
their re-balancing their frequency based on market structure and sector behaviour rather than
applying a one-size-fits-all approach. Although this requires public investors to put in more time
and research to execute, this action does not require investors to have complex understanding and
expertise in market knowledge and investment theories to excel.

6.3. Future Research and Potential Improvements

While the median rotation strategy has shown robustness in multiple markets, several
opportunities exist to enhance the current system and extend its application:

1. Application to Other Markets and Asset Classes

 Testing the median rotation strategy in European indices, such as the FTSE 100, DAX,
or CAC 40, would help assess its effectiveness in different regulatory and economic
environments.
 Expanding to emerging markets, such as the Shanghai Composite or BSE Sensex, could
provide insights into how sector-based rotation performs in higher-growth, higher-
volatility economies.
 Applying median rotation to bond ETFs could determine whether interest rate
sensitivity and credit spreads follow similar mid-tier performance patterns.
 Investigating the median strategy in commodity futures (e.g., oil, gold, and agricultural
products) could reveal whether momentum and mean-reversion effects operate in a
similar manner to equity markets.

2. Data Handling

 A longer time series data would be beneficial for this research in order to ensure that
more market cycles are included, this would improve the robustness of the results
 Future studies could incorporate real time data which would allow dynamic rebalancing
based on real time market performance

3. Multi-Asset Portfolios and Cross-Asset Strategy

10
 International Journal of Data Mining & Knowledge Management Process (IJDKP), Vol.15, No. 3, May 2025

 Future research could examine whether a median-based allocation strategy within a

balanced portfolio (stocks, bonds, real estate, and commodities) outperforms traditional
risk-parity and targetvolatility strategies.

4. Incorporating Machine Learning and Advanced Analytics

 Incorporating factor-based enhancements such as Fama-French factors or

macroeconomic indicators could refine the strategy by adjusting for sector-specific
economic conditions [14] .
 Using machine learning algorithms ([18], [19]) to dynamically adjust the definition of
"median" based on changing market conditions could further improve long-term results.

5. Sector Rotation During Market Crises

 While our study shows that the median strategy performs well during normal market
conditions, future research could analyse its resilience during major economic
downturns (e.g., 2008 financial crisis).

6. Future Expansion

 We understand that we have not covered all aspects of the studies, particularly on
statistical significance testing, transaction cost accounting, and survivorship bias
mitigation. However, we intend to expand on the rotational study series further and will
use future papers to address these concerns

7. CONCLUSION
This study highlights the effectiveness of the median sector rotation strategy in the Nikkei 500
component sectors, reinforcing the allure of Nikkei 500’s ability to balance risk and return better
than winner- or loser-focused strategies. Compared to buy-and-hold investing, the median
approach reduces volatility while maintaining steady long-term growth. Making the median
sector rotation strategy a compelling alternative for public investors.

For public investors, investing in the Japanese sector market via the Nikkei 500 means gaining
exposure to Japan’s most dominant industries while mitigating risk through sector rotation
instead of passive indexing. Our results strongly suggest that the best way to implement this
strategy is through an equal-weighted portfolio of mid-performing sectors, re-balancing the
portfolio quarterly or semi-annually, to optimize risk-adjusted returns.

The strength of the median strategy lies in its simplicity and adaptability, proving that structured,
rule-based investing can have the potential to outperform conventional market approaches. As the
global markets evolve, the ability to strategically rotate between market sectors rather than
passively following an index may become an essential and easy-to-use tool for investors seeking
resilience and long-term performance.

DECLARATIONS
Conflict of Interest: There are no conflicts of interest regarding the publication of this paper.
Author Contributions: All the authors contributed equally to the effort.

11
 International Journal of Data Mining & Knowledge Management Process (IJDKP), Vol.15, No. 3, May 2025

Funding: This research was conducted without any external funding. All aspects of the study,
including design, data collection, analysis, and interpretation, were carried out using the
resources available within the authors’ institution.

Data Availability (including Appendices): All the relevant data, Python code for analysis,
detailed annual tables and graphs are available via:
https://github.com/burotationalpaper/japancomponentsnikkei500

Acknowledgments: The authors would like to thank Metropolitan College of Boston University
for their support.

REFERENCES

[1] M. Isichenko, Quantitative Portfolio Management: The Art and Science of Statistical Arbitrage.
Wiley, 2021, isbn: 978-1-119-82132-8.
[2] J. Miao and P. Polak, Online ensemble of models for optimal predictive performance with
applications to sector rotation strategy, arXiv preprint, 2023. arXiv: 2304.09947 [q-fin.PM].
[Online]. Available: https://arxiv.org/abs/2304.09947.
[3] P. Schanbacher, “Combining portfolio models”, Annals of Economics and Finance, vol. 15, no. 2,
pp. 433–455, 2014. [Online]. Available: http://aeconf.com/Articles/Nov2014/aef150208.pdf.
[4] M. Yang, “Capitalizing on sector rotation strategies”, Technical Analysis of Stocks and Commodities,
vol. 36, no. 2, pp. 22–25, 45, 2018. [Online]. Available: https://store.traders.com/stcov36245ca.html.
[5] E. Pinsky and Y. H. Yang, “A simple rotation strategy with sector etfs”, Technical Analysis of Stocks
and Commodities, vol. 40, no. 12, pp. 36– 40, 2022. Available: https://hdl.handle.net/2144/46846.
[6] S. V. Kumar, S. Valath, and E. Pinsky, “Sell the losers? keep the winners? none of the above – focus
on the median!”, Algorithmic Finance, vol. 10, no. 3–4, pp. 115–135, 2025. doi:
10.1177/21576203241307779.
[7] T. Sharma, A. Srivastava, and E. Pinsky, “Optimizing sector index rotation and rebalancing
frequency with data mining: A case study on indian national stock exchange”, International Journal
of Data Mining & Knowledge Management Process, vol. 15, no. 2, pp. 55–64, 2025. doi:
10.5121/ijdkp.2025.15205
[8] A. L. Johnson, ETF Strategy: Sector Rotation. Wiley, 2024.
[9] T. McIntosh, The Sector Strategist: Using New Asset Allocation Techniques to Reduce Risk and
Improve Investment Returns. Wiley, 2012.
[10] J. Nyaradi, Super Sectors: How to Outsmart the Market Using Sector Rotation. Wiley, 2010.
[11] L. Ma, Nonlinear Investing: A Quantamental Approach. Springer, 2025. doi: 10.1007/978-3-031-
76305-2. [Online]. Available: https://doi.org/10.1007/978-3-031-76305-2.
[12] S. Yoon and H. Takahashi, “Intraday trading dynamics of characteristics and sentiment tendencies of
past news in the tokyo stock exchange market”, Computational Economics, 2024. doi:
10.1007/s10614-02410768-1. [Online]. Available: https://doi.org/10.1007/s10614- 024-10768-1.
[13] M. S. R. Khan, “Hard to borrow vs. easy to borrow: Insights from japan’s centralized lendable stock
market”, International Journal of Financial Studies, vol. 13, no. 1, p. 16, 2025. doi:
10.3390/ijfs13010016. [Online]. Available: https://doi.org/10.3390/ijfs13010016.
[14] E. F. Fama and K. R. French, “Common risk factors in the returns on stocks and bonds”, Journal of
Financial Economics, vol. 33, no. 1, pp. 3–56, 1993. doi: 10.1016/0304-405X(93)90023-5. [Online].
Available: https://doi.org/10.1016/0304-405X(93)90023-5.
[15] K. Tissayakorn, Y. Song, M. Qiu, and F. Akagi, “A study on effectiveness of the “dogs of the dow”
strategy for the thai stock investment”, International Journal of Innovation, Management and
Technology, vol. 4, no. 2, pp. 277–280, 2013. doi: 10.7763/IJIMT.2013.V4.406. [Online]. Available:
https://doi.org/10.7763/IJIMT.2013.V4.406.
[16] A. Beleznay, M. Markov, and A. Panchekha, “Hidden benefits of equal weighting: The case for
hedge fund indices”, SSRN Electronic Journal, Tech. Rep. 1716547, 2010, Posted: 29 Nov 2010;
Last revised: 8 Dec 2011. doi: 10.2139/ssrn.1716547. [Online]. Available: https://doi.
org/10.2139/ssrn.1716547.
[17] N. Jegadeesh and S. Titman, “Returns to buying winners and selling losers: Implications for stock
market efficiency”, The Journal of Finance, vol. 48, no. 1, pp. 65–91, Mar. 1993. doi:
12
 International Journal of Data Mining & Knowledge Management Process (IJDKP), Vol.15, No. 3, May 2025
10.1111/j.15406261.1993.tb04702.x. [Online]. Available: https://doi.org/10.1111/j.1540-
6261.1993.tb04702.x.
[18] C. Bishop, Pattern Recognition and Machine Learning. Springer, 2016.
[19] T. Hastle, Elements of Statistical Learning. Pearson, 2018.

AUTHORS

Sreeya Reddy Kotrakona Harinatha recently completed a Master’s degree in

Applied Data Analytics at Boston University (September 2023 – January 2025). Her
research interests include natural language processing (NLP), predictive models,
forecasting techniques, data analytics, and interactive data visualization. She enjoys
analyzing data to uncover meaningful insights and presenting them effectively. Her
work focuses on leveraging advanced tools and analytical methods to solve complex
problems, identify emerging trends, and support data driven decision-making. She is
particularly interested in the intersection of technology and analytics, where
innovative approaches can reveal impactful solutions and drive better outcomes.

Natasya Liew holds a Bachelor of Arts in Political Science and Economics from
the University of British Columbia, a Master of Science in Applied Data Analytics
from Boston University, and is completing a Master of Science in Computer
Science at Boston University. She has a background in entrepreneurship,
management, and sales, having founded and successfully exited startups in the
AdTech and Healthcare industries. Her work has spanned business development,
technology integration, and scaling operations. In addition to her private-sector
experience, she has been actively involved in the nonprofit sector, leading
fundraising initiatives to support healthcare access in remote regions. Notably, she played a key role in
securing funding and certification for Indonesia’s first floating hospital barge, addressing critical gaps in
medical infrastructure within Indonesia. She also contributed to the World Bank’s 70th Anniversary
Publication on Indonesia Relations, reflecting her broader engagement in policy and economic research.
Her research focuses include optimizing image-to-text models for IoT devices and exploring reinforcement
learning methodologies for improved AI decision-making.

Dr. Eugene Pinsky is an Associate Professor of Practice at Boston University

Metropolitan College. He teaches courses in data science and computational
mathematics for machine learning. His research area is in modeling, performance
evaluations and computational methods in data science and machine learning,
His particular focus is on applications of machine learning and data mining to
algorithmic trading, pricing models, financial modeling, risk, and portfolio
construction.

APPENDIX A: GROWTH COMPARISON FOR DIFFERENT ROTATION FREQUENCIES

13
International Journal of Data Mining & Knowledge Management Process (IJDKP), Vol.15, No. 3, May 2025
Figure 5: Growth Comparison with Weekly Re-balancing

Figure 6: Growth Comparison with Monthly Re-balancing

Figure 7: Growth Comparison with Quarterly Re-balancing

Figure 8: Growth Comparison with Half-Year Strategy

14
International Journal of Data Mining & Knowledge Management Process (IJDKP), Vol.15, No. 3, May 2025

Figure 9: Growth Comparison with Annual Re-balancing

15