Factor Strategy – Idiosyncratic Volatility 

Introduction 

Building on the statistical foundation presented in Part 1, this article explores how Idiosyncratic Volatility (IVOL) can be effectively applied in investment strategy design. We present two categories of approaches: a single-factor sorting model and a set of filter-enhanced momentum strategies. Through robust backtesting across two decades of Taiwan stock market data, we demonstrate how IVOL can improve risk-adjusted performance when used as a portfolio filter—especially when combined with momentum or dividend-based signals. 

👉 Part 1：Factor Research－Idiosyncratic Volatility

Strategy Construction & Backtesting  

Building on the earlier analysis, we now evaluate IVOL’s practical investment applications. Two categories of strategies are tested: 

1. Filter-Enhanced Momentum Strategies – incorporating IVOL and/or DP as pre-filters to refine the MOM52WH signal. 
2. Single-Factor Sorting Strategies – directly using IVOL, dividend yield (DP), or momentum (MOM52WH) as the basis for monthly stock selection; 

Backtests are conducted over Jan 2005 – Mar 2025, rebalanced monthly, with an initial capital of NT$10 million allocated equally across 50 selected stocks. 

Single-Factor Sorting Strategies 

Each month, stocks are ranked by one of the following factors, and the top 50 are selected: 

• IVOL: bottom 50 (lowest volatility) 
• MOM52WH: top 50 stocks near 52-week highs 
• DP: top 50 by cash dividend yield 

📌 All data and factors are sourced from the TEJ Factor Library, which enables clean monthly signal construction and portfolio simulation. 

Table 1: Performance of Single-Factor Strategies (

Benchmark: TAIEX Total Return Index, Jan 2005 – Mar 2025) 

Metric	IVOL	MOM52WH	DP	Benchmark
Annual Return	3.57%	13.40%	5.93%	10.41%
Cumulative Return	99.93%	1099.27%	211.92%	607.56%
Annual Volatility	7.94%	12.27%	12.96%	18.02%
Sharpe Ratio	0.482	1.087	0.510	0.640
Sortino Ratio	0.635	1.490	0.670	0.887
Max Drawdown	-38.72%	-37.70%	-50.49%	-56.02%
Alpha	-0.001	0.082	0.002	–
Beta	0.343	0.470	0.559	–

Figure 2: Cumulative Returns of Single-Factor Strategies 

Insights 

• IVOL alone provides limited return (3.57%), but its low volatility (7.94%) and low beta (0.34) make it suitable as a defensive anchor in portfolio construction. 
• MOM52WH delivers the highest alpha and Sharpe ratio, confirming momentum’s strength in Taiwan. 
• DP strategy shows moderate returns but high drawdowns—indicating dividend stocks are not always stable. 

💡 Conclusion: IVOL is not ideal as a standalone alpha factor, but it has potential when used in combination with other signals for risk control. 

Filter-Enhanced Momentum Strategies

To improve momentum strategies, we apply IVOL and/or DP as filters before ranking by MOM52WH: 

• IVOL Filter: exclude top 50% IVOL stocks, then select top 50 MOM52WH. 
• DP Filter: exclude bottom 50% dividend yield stocks, then select top MOM52WH. 
• IVOL + DP Filter: exclude both high-IVOL and low-DP stocks, then select MOM52WH top 50. 

This approach reflects prior findings—high-IVOL stocks often have low dividend yield and poor return quality. 

Table 3: Performance of Filter-Enhanced Momentum Strategies (Jan 2005 – Mar 2025) 

Metric	MOM52WH	+ IVOL Filter	+ DP Filter	+ IVOL & DP Filters	Benchmark
Annual Return	13.40%	12.93%	15.23%	15.44%	10.41%
Cumulative Return	1099.27%	1004.23%	1542.76%	1603.16%	607.56%
Annual Volatility	12.27%	8.79%	9.50%	8.68%	18.02%
Sharpe Ratio	1.087	1.428	1.540	1.697	0.640
Sortino Ratio	1.490	1.964	2.123	2.346	0.887
Max Drawdown	-37.70%	-37.81%	-36.75%	-36.72%	-56.02%
Alpha	0.082	0.084	0.103	0.108	–
Beta	0.470	0.390	0.418	0.389	–

Figure 4: Cumulative Returns of Filtered Strategies 

Interpretation 

• Filtering by IVOL reduces volatility significantly, with only slight impact on return. 
• DP filter improves both return and Sharpe ratio, suggesting dividend yield complements momentum well. 
• The IVOL + DP filter strategy performs best overall, combining strong returns with lower risk and minimal drawdown. 

📌 This validates IVOL’s value as a risk management filter, especially when paired with alpha-generating signals like momentum. 

Conclusion 

This study examines the validity of Idiosyncratic Volatility (IVOL) as an investment factor in Taiwan’s stock market. After reviewing theoretical foundations and prior research, we first confirmed the presence of the low-volatility anomaly in Taiwan. We then analyzed the cross-sectional relationship between IVOL and future returns. 

The factor analysis revealed a stable and significant negative relationship between IVOL and subsequent stock performance. However, when IVOL was used alone to construct a single-factor strategy, the resulting returns were relatively weak. In contrast, using IVOL as a risk filter in combination with momentum factors led to substantial improvements in risk-adjusted returns, suggesting that excluding high-IVOL stocks may enhance overall portfolio quality. 

Overall, IVOL may not be suitable as a standalone selection factor, but it holds practical value when used as a risk screening condition. This conclusion aligns with the view that low-volatility anomalies are driven by the underperformance of high-risk stocks, rather than the superior returns of low-risk stocks. 

In practice, IVOL supports not only return enhancement, but also risk control and portfolio construction in a multi-factor setting. 

💼 Empower Your Quant Strategy with TEJ 

The TEJ Factor Library provides over 100 standardized alpha factors—including IVOL, momentum, valuation, and quality—built with point-in-time methodology for clean backtests. Combined with TEJ Market & Financial Data, it enables seamless: 

📩Explore how IVOL and other factors perform in your portfolio universe? 
Visit TEJ Factor Library or contact us to request a custom factor analysis demo. 

Further Reading:

• How Dividend Policy Affects Investment: An Event Study Analysis of Key Factors
• Charles Brandes’ Value Investing Principles : Building a Portfolio with a Margin of Safety