{"id":39409,"date":"2025-08-12T17:00:00","date_gmt":"2025-08-12T09:00:00","guid":{"rendered":"https:\/\/www.tejwin.com\/?post_type=insight&#038;p=39409"},"modified":"2025-09-08T15:13:34","modified_gmt":"2025-09-08T07:13:34","slug":"factor-research-capital-gain-overhang-part-1","status":"publish","type":"insight","link":"https:\/\/www.tejwin.com\/en\/insight\/factor-research-capital-gain-overhang-part-1\/","title":{"rendered":"Factor Research \u2013Capital Gain Overhang | Part 1"},"content":{"rendered":"\n<figure class=\"wp-block-image size-large\"><img fetchpriority=\"high\" decoding=\"async\" width=\"1024\" height=\"576\" src=\"https:\/\/www.tejwin.com\/wp-content\/uploads\/\/Factor-Research_CGO-1024x576.png\" alt=\"\" class=\"wp-image-39411\" srcset=\"https:\/\/www.tejwin.com\/wp-content\/uploads\/Factor-Research_CGO-1024x576.png 1024w, https:\/\/www.tejwin.com\/wp-content\/uploads\/Factor-Research_CGO-300x169.png 300w, https:\/\/www.tejwin.com\/wp-content\/uploads\/Factor-Research_CGO-150x84.png 150w, https:\/\/www.tejwin.com\/wp-content\/uploads\/Factor-Research_CGO-768x432.png 768w, https:\/\/www.tejwin.com\/wp-content\/uploads\/Factor-Research_CGO-1536x864.png 1536w, https:\/\/www.tejwin.com\/wp-content\/uploads\/Factor-Research_CGO.png 1920w\" sizes=\"(max-width: 1024px) 100vw, 1024px\" \/><\/figure>\n\n\n\n<div id=\"ez-toc-container\" class=\"ez-toc-v2_0_81 counter-hierarchy ez-toc-counter ez-toc-grey ez-toc-container-direction\">\n<p class=\"ez-toc-title\" style=\"cursor:inherit\">Table of Contents<\/p>\n<label for=\"ez-toc-cssicon-toggle-item-69f0e5d857ff8\" class=\"ez-toc-cssicon-toggle-label\"><span class=\"ez-toc-cssicon\"><span class=\"eztoc-hide\" style=\"display:none;\">Toggle<\/span><span class=\"ez-toc-icon-toggle-span\"><svg style=\"fill: #999;color:#999\" xmlns=\"http:\/\/www.w3.org\/2000\/svg\" class=\"list-377408\" width=\"20px\" height=\"20px\" viewBox=\"0 0 24 24\" fill=\"none\"><path d=\"M6 6H4v2h2V6zm14 0H8v2h12V6zM4 11h2v2H4v-2zm16 0H8v2h12v-2zM4 16h2v2H4v-2zm16 0H8v2h12v-2z\" fill=\"currentColor\"><\/path><\/svg><svg style=\"fill: #999;color:#999\" class=\"arrow-unsorted-368013\" xmlns=\"http:\/\/www.w3.org\/2000\/svg\" width=\"10px\" height=\"10px\" viewBox=\"0 0 24 24\" version=\"1.2\" baseProfile=\"tiny\"><path d=\"M18.2 9.3l-6.2-6.3-6.2 6.3c-.2.2-.3.4-.3.7s.1.5.3.7c.2.2.4.3.7.3h11c.3 0 .5-.1.7-.3.2-.2.3-.5.3-.7s-.1-.5-.3-.7zM5.8 14.7l6.2 6.3 6.2-6.3c.2-.2.3-.5.3-.7s-.1-.5-.3-.7c-.2-.2-.4-.3-.7-.3h-11c-.3 0-.5.1-.7.3-.2.2-.3.5-.3.7s.1.5.3.7z\"\/><\/svg><\/span><\/span><\/label><input type=\"checkbox\"  id=\"ez-toc-cssicon-toggle-item-69f0e5d857ff8\"  aria-label=\"Toggle\" \/><nav><ul class='ez-toc-list ez-toc-list-level-1 ' ><li class='ez-toc-page-1 ez-toc-heading-level-2'><a class=\"ez-toc-link ez-toc-heading-1\" href=\"https:\/\/www.tejwin.com\/en\/insight\/factor-research-capital-gain-overhang-part-1\/#Preface\" >Preface<\/a><\/li><li class='ez-toc-page-1 ez-toc-heading-level-2'><a class=\"ez-toc-link ez-toc-heading-2\" href=\"https:\/\/www.tejwin.com\/en\/insight\/factor-research-capital-gain-overhang-part-1\/#Capital_Gain_Overhang_CGO_A_Key_Measure_of_the_Disposition_Effect\" >Capital Gain Overhang (CGO) : A Key Measure of the Disposition Effect<\/a><\/li><li class='ez-toc-page-1 ez-toc-heading-level-2'><a class=\"ez-toc-link ez-toc-heading-3\" href=\"https:\/\/www.tejwin.com\/en\/insight\/factor-research-capital-gain-overhang-part-1\/#Factor_Analysis\" >Factor Analysis<\/a><ul class='ez-toc-list-level-3' ><li class='ez-toc-heading-level-3'><a class=\"ez-toc-link ez-toc-heading-4\" href=\"https:\/\/www.tejwin.com\/en\/insight\/factor-research-capital-gain-overhang-part-1\/#Data_Source_and_Sample_Period\" >Data Source and Sample Period<\/a><\/li><li class='ez-toc-page-1 ez-toc-heading-level-3'><a class=\"ez-toc-link ez-toc-heading-5\" href=\"https:\/\/www.tejwin.com\/en\/insight\/factor-research-capital-gain-overhang-part-1\/#Construction_of_Capital_Gain_Overhang_CGO\" >Construction of Capital Gain Overhang (CGO)<\/a><\/li><\/ul><\/li><li class='ez-toc-page-1 ez-toc-heading-level-2'><a class=\"ez-toc-link ez-toc-heading-6\" href=\"https:\/\/www.tejwin.com\/en\/insight\/factor-research-capital-gain-overhang-part-1\/#Descriptive_Statistics\" >Descriptive Statistics<\/a><\/li><li class='ez-toc-page-1 ez-toc-heading-level-2'><a class=\"ez-toc-link ez-toc-heading-7\" href=\"https:\/\/www.tejwin.com\/en\/insight\/factor-research-capital-gain-overhang-part-1\/#Return_Analysis\" >Return Analysis<\/a><\/li><li class='ez-toc-page-1 ez-toc-heading-level-2'><a class=\"ez-toc-link ez-toc-heading-8\" href=\"https:\/\/www.tejwin.com\/en\/insight\/factor-research-capital-gain-overhang-part-1\/#Risk_Factor_Model_Regression\" >Risk Factor Model Regression<\/a><\/li><li class='ez-toc-page-1 ez-toc-heading-level-2'><a class=\"ez-toc-link ez-toc-heading-9\" href=\"https:\/\/www.tejwin.com\/en\/insight\/factor-research-capital-gain-overhang-part-1\/#Information_Coefficient_IC_Analysis\" >Information Coefficient (IC) Analysis<\/a><\/li><li class='ez-toc-page-1 ez-toc-heading-level-2'><a class=\"ez-toc-link ez-toc-heading-10\" href=\"https:\/\/www.tejwin.com\/en\/insight\/factor-research-capital-gain-overhang-part-1\/#Conclusion\" >Conclusion<\/a><\/li><\/ul><\/nav><\/div>\n<h2 class=\"wp-block-heading\"><span class=\"ez-toc-section\" id=\"Preface\"><\/span><strong>Preface<\/strong><span class=\"ez-toc-section-end\"><\/span><\/h2>\n\n\n\n<p>The origins of the momentum anomaly have long been debated, with multiple competing explanations. Among them, one of the most influential behavioral interpretations attributes momentum to the <strong><mark style=\"background-color:rgba(0, 0, 0, 0)\" class=\"has-inline-color has-luminous-vivid-orange-color\">Disposition Effect<\/mark>,<\/strong> a systematic bias in investor decision-making. This article focuses on the <strong><mark style=\"background-color:rgba(0, 0, 0, 0)\" class=\"has-inline-color has-luminous-vivid-orange-color\">Capital Gain Overhang (CGO)<\/mark><\/strong> factor, specifically designed to quantify this behavioral bias. Using the Taiwan equity market as a case study, we examine CGO\u2019s predictive power as a stock selection indicator and evaluate its practical value through empirical analysis.<\/p>\n\n\n\n<h2 class=\"wp-block-heading\"><span class=\"ez-toc-section\" id=\"Capital_Gain_Overhang_CGO_A_Key_Measure_of_the_Disposition_Effect\"><\/span>Capital Gain Overhang (CGO) : A Key Measure of the<br>Disposition Effect<span class=\"ez-toc-section-end\"><\/span><\/h2>\n\n\n\n<p>A cornerstone of behavioral finance is Prospect Theory, proposed by Kahneman and Tversky (1979). The theory suggests that individuals evaluate gains and losses relative to a <strong><mark style=\"background-color:rgba(0, 0, 0, 0)\" class=\"has-inline-color has-luminous-vivid-orange-color\">reference point<\/mark><\/strong> rather than absolute wealth levels, and they experience losses more intensely than equivalent gains\u2014a phenomenon known as<strong> <\/strong><mark style=\"background-color:rgba(0, 0, 0, 0)\" class=\"has-inline-color has-luminous-vivid-orange-color\"><strong>loss aversion<\/strong>.<\/mark><\/p>\n\n\n\n<p>Building on this foundation, Shefrin and Statman (1985) introduced the concept of the Disposition Effect. Within the framework of <strong><mark style=\"background-color:rgba(0, 0, 0, 0)\" class=\"has-inline-color has-luminous-vivid-orange-color\">mental accounting<\/mark><\/strong>, investors tend to assign each stock purchase to a separate mental account, with the purchase price serving as the critical reference point for gains or losses. This bias drives two systematic behaviors: investors are prone to selling winning stocks too early to lock in gains, while holding onto losing stocks for too long to avoid realizing losses.<\/p>\n\n\n\n<p>If such behavior is widespread, it inevitably creates predictable price pressure. To capture this effect, Grinblatt and Han (2005) proposed the Capital Gain Overhang (CGO) factor, designed to directly quantify the Disposition Effect.<mark style=\"background-color:rgba(0, 0, 0, 0)\" class=\"has-inline-color has-luminous-vivid-orange-color\"> <strong>CGO measures the gap between the current market price and the estimated average cost basis of all shareholders.<\/strong> <\/mark>Since actual investor costs are unobservable, they introduced a turnover-based method, which approximates cost using a volume-weighted series of historical prices. Subsequent research validated this approach: Frazzini (2006) employed a completely different <strong><mark style=\"background-color:rgba(0, 0, 0, 0)\" class=\"has-inline-color has-luminous-vivid-orange-color\">holding-based method<\/mark><\/strong> using mutual fund holdings data and reached conclusions highly consistent with Grinblatt and Han. The convergence of these independent methodologies strongly indicates that <strong><mark style=\"background-color:rgba(0, 0, 0, 0)\" class=\"has-inline-color has-luminous-vivid-orange-color\">the CGO effect is genuine<\/mark><\/strong>, rather than an artifact of any specific estimation technique.<\/p>\n\n\n\n<p>The theoretical implication is straightforward:<\/p>\n\n\n\n<ul class=\"wp-block-list\">\n<li>When most investors hold unrealized gains (high CGO), potential selling pressure from profit-taking dampens the market\u2019s reaction to positive news.<\/li>\n\n\n\n<li>Conversely, when investors hold unrealized losses (low CGO), their reluctance to realize losses provides support, muting the impact of negative news.<\/li>\n<\/ul>\n\n\n\n<p>This asymmetric reaction creates<strong><mark style=\"background-color:rgba(0, 0, 0, 0)\" class=\"has-inline-color has-luminous-vivid-orange-color\"> a positive relationship between CGO and expected future returns.<\/mark><\/strong><\/p>\n\n\n\n<p>Most importantly, CGO reinterprets the traditional momentum factor. Grinblatt and Han (2005) demonstrated that CGO not only predicts future returns but also subsumes the explanatory power of the conventional intermediate-term momentum factor. This suggests that the well-documented momentum anomaly may largely be a manifestation of the Disposition Effect. In other words, <strong><mark style=\"background-color:rgba(0, 0, 0, 0)\" class=\"has-inline-color has-luminous-vivid-orange-color\">traditional momentum factors, based purely on past returns, may simply act as a noisy proxy for CGO<\/mark><\/strong>.<\/p>\n\n\n\n<figure class=\"wp-block-image size-large\"><a href=\"https:\/\/www.tejwin.com\/en\/news\/factor-library\/\"><img decoding=\"async\" width=\"1024\" height=\"107\" src=\"https:\/\/www.tejwin.com\/wp-content\/uploads\/\/CTA_Factor-Library-1-1024x107.png\" alt=\"\" class=\"wp-image-35334\" srcset=\"https:\/\/www.tejwin.com\/wp-content\/uploads\/CTA_Factor-Library-1-1024x107.png 1024w, https:\/\/www.tejwin.com\/wp-content\/uploads\/CTA_Factor-Library-1-300x31.png 300w, https:\/\/www.tejwin.com\/wp-content\/uploads\/CTA_Factor-Library-1-150x16.png 150w, https:\/\/www.tejwin.com\/wp-content\/uploads\/CTA_Factor-Library-1-768x80.png 768w, https:\/\/www.tejwin.com\/wp-content\/uploads\/CTA_Factor-Library-1-1536x160.png 1536w, https:\/\/www.tejwin.com\/wp-content\/uploads\/CTA_Factor-Library-1.png 1920w\" sizes=\"(max-width: 1024px) 100vw, 1024px\" \/><\/a><\/figure>\n\n\n\n\n\n<h2 class=\"wp-block-heading\"><span class=\"ez-toc-section\" id=\"Factor_Analysis\"><\/span>Factor Analysis<span class=\"ez-toc-section-end\"><\/span><\/h2>\n\n\n\n<p>This section presents the empirical examination of the Capital Gain Overhang (CGO) factor in Taiwan\u2019s equity market, focusing on whether it reliably predicts future returns.<\/p>\n\n\n\n<h3 class=\"wp-block-heading\"><span class=\"ez-toc-section\" id=\"Data_Source_and_Sample_Period\"><\/span>Data Source and Sample Period<span class=\"ez-toc-section-end\"><\/span><\/h3>\n\n\n\n<p>The dataset is sourced from Taiwan Economic Journal (TEJ) :<\/p>\n\n\n\n<ul class=\"wp-block-list\">\n<li>Price &amp; Trading Data: stock price and returns.<\/li>\n\n\n\n<li>Market Factor Data : market risk premium, size premium, book-to-market premium, and risk-free rate<\/li>\n\n\n\n<li>Factor Indicators : Capital Gain Overhang (CGO), 52-week high momentum (MOM52WH)<\/li>\n\n\n\n<li>Sample period: Jan 2005 \u2013 June 2025<\/li>\n\n\n\n<li>Scope: All common stocks listed on the Taiwan Stock Exchange (TWSE) and Taipei Exchange (TPEx).<\/li>\n<\/ul>\n\n\n\n<h3 class=\"wp-block-heading\"><span class=\"ez-toc-section\" id=\"Construction_of_Capital_Gain_Overhang_CGO\"><\/span>Construction of Capital Gain Overhang (CGO)<span class=\"ez-toc-section-end\"><\/span><\/h3>\n\n\n\n<p>CGO is calculated following Grinblatt and Han\u2019s (2005) turnover-based method, adapted to Taiwan\u2019s market structure. Specifically, TEJ applies a 100-trading-day look-back window, using<strong> turnover-weighted adjusted daily prices<\/strong> with time decay to estimate investors\u2019 average cost basis.<\/p>\n\n\n\n<p>This design reflects two key assumptions:<\/p>\n\n\n\n<ol class=\"wp-block-list\">\n<li><strong>Turnover weighting<\/strong> \u2013 high-turnover days better capture shifts in the cost base.<\/li>\n\n\n\n<li><strong>Time decay<\/strong> \u2013 older prices are less representative due to position turnover, consistent with investors\u2019 mental accounting.<\/li>\n<\/ol>\n\n\n\n<h2 class=\"wp-block-heading\"><span class=\"ez-toc-section\" id=\"Descriptive_Statistics\"><\/span><a>Descriptive Statistics<\/a><span class=\"ez-toc-section-end\"><\/span><\/h2>\n\n\n\n<p>To examine the cross-sectional features of Capital Gain Overhang (CGO) in Taiwan\u2019s stock market, we apply the <strong>portfolio-sorting method<\/strong>. On each trading day, all stocks are ranked by their CGO values and divided into ten equally weighted portfolios, labeled from P1 (lowest CGO) to P10 (highest CGO).<\/p>\n\n\n\n<p>Table 1 summarizes descriptive statistics of these portfolios over the sample period (January 2005 \u2013 June 2025). The results show that average CGO values increase monotonically from P1 to P10, confirming the effectiveness of the grouping.<\/p>\n\n\n\n<p>Table 1\uff1adescriptive statistics of CGO factor<\/p>\n\n\n\n<figure class=\"wp-block-table is-style-stripes\"><table class=\"has-background\" style=\"background-color:#ffe9ae\"><thead><tr><th><\/th><th class=\"has-text-align-right\" data-align=\"right\">Min<\/th><th class=\"has-text-align-right\" data-align=\"right\">Max<\/th><th class=\"has-text-align-right\" data-align=\"right\">Mean<\/th><th class=\"has-text-align-right\" data-align=\"right\">Std<\/th><th class=\"has-text-align-right\" data-align=\"right\">Count<\/th><th class=\"has-text-align-right\" data-align=\"right\"> %<\/th><\/tr><\/thead><tbody><tr><td>1<\/td><td class=\"has-text-align-right\" data-align=\"right\">-0.9236<\/td><td class=\"has-text-align-right\" data-align=\"right\">0.09118<\/td><td class=\"has-text-align-right\" data-align=\"right\">-0.194762<\/td><td class=\"has-text-align-right\" data-align=\"right\"><strong>0.107928<\/strong><\/td><td class=\"has-text-align-right\" data-align=\"right\">735360<\/td><td class=\"has-text-align-right\" data-align=\"right\">10.031237<\/td><\/tr><tr><td>2<\/td><td class=\"has-text-align-right\" data-align=\"right\">-0.53369<\/td><td class=\"has-text-align-right\" data-align=\"right\">0.12821<\/td><td class=\"has-text-align-right\" data-align=\"right\">-0.114891<\/td><td class=\"has-text-align-right\" data-align=\"right\">0.076839<\/td><td class=\"has-text-align-right\" data-align=\"right\">732881<\/td><td class=\"has-text-align-right\" data-align=\"right\">9.99742<\/td><\/tr><tr><td>3<\/td><td class=\"has-text-align-right\" data-align=\"right\">-0.46608<\/td><td class=\"has-text-align-right\" data-align=\"right\">0.15705<\/td><td class=\"has-text-align-right\" data-align=\"right\">-0.081821<\/td><td class=\"has-text-align-right\" data-align=\"right\">0.07067<\/td><td class=\"has-text-align-right\" data-align=\"right\">732350<\/td><td class=\"has-text-align-right\" data-align=\"right\">9.990177<\/td><\/tr><tr><td>4<\/td><td class=\"has-text-align-right\" data-align=\"right\">-0.42359<\/td><td class=\"has-text-align-right\" data-align=\"right\">0.18262<\/td><td class=\"has-text-align-right\" data-align=\"right\">-0.057136<\/td><td class=\"has-text-align-right\" data-align=\"right\">0.065729<\/td><td class=\"has-text-align-right\" data-align=\"right\">732840<\/td><td class=\"has-text-align-right\" data-align=\"right\">9.996861<\/td><\/tr><tr><td>5<\/td><td class=\"has-text-align-right\" data-align=\"right\">-0.38765<\/td><td class=\"has-text-align-right\" data-align=\"right\">0.20536<\/td><td class=\"has-text-align-right\" data-align=\"right\">-0.036038<\/td><td class=\"has-text-align-right\" data-align=\"right\">0.061512<\/td><td class=\"has-text-align-right\" data-align=\"right\">733393<\/td><td class=\"has-text-align-right\" data-align=\"right\">10.004405<\/td><\/tr><tr><td>6<\/td><td class=\"has-text-align-right\" data-align=\"right\">-0.34851<\/td><td class=\"has-text-align-right\" data-align=\"right\">0.23364<\/td><td class=\"has-text-align-right\" data-align=\"right\">-0.016373<\/td><td class=\"has-text-align-right\" data-align=\"right\">0.057915<\/td><td class=\"has-text-align-right\" data-align=\"right\">731802<\/td><td class=\"has-text-align-right\" data-align=\"right\">9.982702<\/td><\/tr><tr><td>7<\/td><td class=\"has-text-align-right\" data-align=\"right\">-0.31608<\/td><td class=\"has-text-align-right\" data-align=\"right\">0.27171<\/td><td class=\"has-text-align-right\" data-align=\"right\">0.003678<\/td><td class=\"has-text-align-right\" data-align=\"right\">0.055108<\/td><td class=\"has-text-align-right\" data-align=\"right\">732338<\/td><td class=\"has-text-align-right\" data-align=\"right\">9.990013<\/td><\/tr><tr><td>8<\/td><td class=\"has-text-align-right\" data-align=\"right\">-0.27466<\/td><td class=\"has-text-align-right\" data-align=\"right\">0.3223<\/td><td class=\"has-text-align-right\" data-align=\"right\">0.026796<\/td><td class=\"has-text-align-right\" data-align=\"right\">0.053688<\/td><td class=\"has-text-align-right\" data-align=\"right\">732727<\/td><td class=\"has-text-align-right\" data-align=\"right\">9.99532<\/td><\/tr><tr><td>9<\/td><td class=\"has-text-align-right\" data-align=\"right\">-0.22777<\/td><td class=\"has-text-align-right\" data-align=\"right\">0.40388<\/td><td class=\"has-text-align-right\" data-align=\"right\">0.05915<\/td><td class=\"has-text-align-right\" data-align=\"right\">0.055491<\/td><td class=\"has-text-align-right\" data-align=\"right\">732306<\/td><td class=\"has-text-align-right\" data-align=\"right\">9.989577<\/td><\/tr><tr><td>10<\/td><td class=\"has-text-align-right\" data-align=\"right\">-0.16402<\/td><td class=\"has-text-align-right\" data-align=\"right\">3.2241<\/td><td class=\"has-text-align-right\" data-align=\"right\">0.151017<\/td><td class=\"has-text-align-right\" data-align=\"right\"><strong>0.110712<\/strong><\/td><td class=\"has-text-align-right\" data-align=\"right\">734704<\/td><td class=\"has-text-align-right\" data-align=\"right\">10.022288<\/td><\/tr><\/tbody><\/table><figcaption class=\"wp-element-caption\"><em>Sample\uff1aAll TSE and OTC common stocks in Taiwan <br>Data period: Jan 2005 \u2013 June 2025<\/em><\/figcaption><\/figure>\n\n\n\n<div style=\"height:25px\" aria-hidden=\"true\" class=\"wp-block-spacer\"><\/div>\n\n\n\n<p>An important observation is that the extreme portfolios\u2014P1 (largest unrealized losers) and P10 (largest unrealized winners)\u2014exhibit significantly higher standard deviations than the middle groups. This indicates greater heterogeneity among firms at both ends of the CGO distribution.<\/p>\n\n\n\n<h2 class=\"wp-block-heading\"><span class=\"ez-toc-section\" id=\"Return_Analysis\"><\/span>Return Analysis<span class=\"ez-toc-section-end\"><\/span><\/h2>\n\n\n\n<p>We next evaluate the predictive power of CGO for future returns by analyzing the performance of these sorted portfolios.<\/p>\n\n\n\n<p>Table 2 and Figure 3 present the average daily returns of CGO decile portfolios across different holding horizons (1 to 100 trading days). Two key findings emerge:<\/p>\n\n\n\n<ol class=\"wp-block-list\">\n<li><strong><mark style=\"background-color:rgba(0, 0, 0, 0)\" class=\"has-inline-color has-luminous-vivid-orange-color\">Return predictability<\/mark><\/strong> \u2013 Across all horizons, the highest-CGO portfolio (P10) consistently outperforms the lowest-CGO portfolio (P1). As a result, the long\u2013short spread (P10\u2013P1) remains positive, confirming that high-CGO stocks tend to deliver higher subsequent returns.<br><\/li>\n\n\n\n<li><mark style=\"background-color:rgba(0, 0, 0, 0)\" class=\"has-inline-color has-luminous-vivid-orange-color\"><strong>Evolution of return patterns<\/strong> <\/mark>\u2013The shape of the return\u2013CGO relationship shifts with the investment horizon.\n<ul class=\"wp-block-list\">\n<li><span style=\"color: initial;\">For short horizons (1\u201340 days), the return curve shows a U-shaped distribution, where both extreme portfolios (P1 and P10) outperform the middle deciles.<\/span><\/li>\n\n\n\n<li><mark style=\"background-color:rgba(0, 0, 0, 0)\" class=\"has-inline-color has-luminous-vivid-orange-color\"><span style=\"background-color: rgba(0, 0, 0, 0.2); color: initial;\">Over longer horizons (60\u2013100 days), the pattern transitions into a monotonically increasing relationship, providing strong empirical support for the hypothesis that <\/span><strong>higher CGO stocks deliver higher long-term expected returns<\/strong>.<\/mark><\/li>\n<\/ul>\n<\/li>\n<\/ol>\n\n\n\n<p>Table 2\uff1aAverage Daily Returns of CGO Factor Groups and Long-Short Hedge Portfolios<\/p>\n\n\n\n<figure class=\"wp-block-table is-style-stripes\"><table class=\"has-background\" style=\"background-color:#ffe9ae\"><thead><tr><th class=\"has-text-align-center\" data-align=\"center\"><\/th><th class=\"has-text-align-right\" data-align=\"right\">1D<\/th><th class=\"has-text-align-right\" data-align=\"right\">5D<\/th><th class=\"has-text-align-right\" data-align=\"right\">10D<\/th><th class=\"has-text-align-right\" data-align=\"right\">20D<\/th><th class=\"has-text-align-right\" data-align=\"right\">40D<\/th><th class=\"has-text-align-right\" data-align=\"right\">60D<\/th><th class=\"has-text-align-right\" data-align=\"right\">100D<\/th><\/tr><\/thead><tbody><tr><td class=\"has-text-align-center\" data-align=\"center\">Top Quantile, P10<\/td><td class=\"has-text-align-right\" data-align=\"right\">9.964<\/td><td class=\"has-text-align-right\" data-align=\"right\">10.178<\/td><td class=\"has-text-align-right\" data-align=\"right\">10.275<\/td><td class=\"has-text-align-right\" data-align=\"right\">8.820<\/td><td class=\"has-text-align-right\" data-align=\"right\">8.293<\/td><td class=\"has-text-align-right\" data-align=\"right\">8.226<\/td><td class=\"has-text-align-right\" data-align=\"right\">8.026<\/td><\/tr><tr><td class=\"has-text-align-center\" data-align=\"center\">Bottom Quantile, P1<\/td><td class=\"has-text-align-right\" data-align=\"right\">7.626<\/td><td class=\"has-text-align-right\" data-align=\"right\">7.322<\/td><td class=\"has-text-align-right\" data-align=\"right\">6.486<\/td><td class=\"has-text-align-right\" data-align=\"right\">6.465<\/td><td class=\"has-text-align-right\" data-align=\"right\">5.387<\/td><td class=\"has-text-align-right\" data-align=\"right\">4.279<\/td><td class=\"has-text-align-right\" data-align=\"right\">4.046<\/td><\/tr><tr><td class=\"has-text-align-center\" data-align=\"center\">Spread, P10-P1<\/td><td class=\"has-text-align-right\" data-align=\"right\">2.338<\/td><td class=\"has-text-align-right\" data-align=\"right\">3.386<\/td><td class=\"has-text-align-right\" data-align=\"right\">4.400<\/td><td class=\"has-text-align-right\" data-align=\"right\">3.010<\/td><td class=\"has-text-align-right\" data-align=\"right\">3.552<\/td><td class=\"has-text-align-right\" data-align=\"right\">4.543<\/td><td class=\"has-text-align-right\" data-align=\"right\">4.541<\/td><\/tr><\/tbody><\/table><figcaption class=\"wp-element-caption\">Unit\uff1abps    Data period: Jan 2005 \u2013 June 2025<\/figcaption><\/figure>\n\n\n\n<div style=\"height:25px\" aria-hidden=\"true\" class=\"wp-block-spacer\"><\/div>\n\n\n\n<p>Figure 3\uff1aAverage returns of CGO factor deciles over different holding periods<\/p>\n\n\n\n<figure class=\"wp-block-image size-full is-resized\"><img decoding=\"async\" width=\"600\" height=\"198\" src=\"https:\/\/www.tejwin.com\/wp-content\/uploads\/2508_CGO1.png\" alt=\"\" class=\"wp-image-38836\" style=\"width:840px;height:auto\" srcset=\"https:\/\/www.tejwin.com\/wp-content\/uploads\/2508_CGO1.png 600w, https:\/\/www.tejwin.com\/wp-content\/uploads\/2508_CGO1-300x99.png 300w, https:\/\/www.tejwin.com\/wp-content\/uploads\/2508_CGO1-150x50.png 150w\" sizes=\"(max-width: 600px) 100vw, 600px\" \/><figcaption class=\"wp-element-caption\">Unit\uff1abps   Data period: Jan 2005 \u2013 June 2025<\/figcaption><\/figure>\n\n\n\n<div style=\"height:25px\" aria-hidden=\"true\" class=\"wp-block-spacer\"><\/div>\n\n\n\n<p>Figure 4 further illustrates cumulative performance. The high-CGO portfolio shows steadily superior cumulative returns compared to the low-CGO portfolio, with smaller drawdowns, suggesting that <mark style=\"background-color:rgba(0, 0, 0, 0)\" class=\"has-inline-color has-luminous-vivid-orange-color\"><strong>high-CGO firms may possess a degree of defensiveness in adverse markets<\/strong>. <\/mark>This feature enhances the factor\u2019s ability to generate favorable risk-adjusted returns over the long run.<\/p>\n\n\n\n<div style=\"height:25px\" aria-hidden=\"true\" class=\"wp-block-spacer\"><\/div>\n\n\n\n<p>Figure 4 : Cumulative Return Trends by CGO Group (Top vs. Bottom Quantile)<\/p>\n\n\n\n<figure class=\"wp-block-image size-large is-resized\"><img loading=\"lazy\" decoding=\"async\" width=\"1024\" height=\"325\" src=\"https:\/\/www.tejwin.com\/wp-content\/uploads\/\/2508_CGO2-1024x325.png\" alt=\"\" class=\"wp-image-38838\" style=\"width:840px;height:auto\" srcset=\"https:\/\/www.tejwin.com\/wp-content\/uploads\/2508_CGO2-1024x325.png 1024w, https:\/\/www.tejwin.com\/wp-content\/uploads\/2508_CGO2-300x95.png 300w, https:\/\/www.tejwin.com\/wp-content\/uploads\/2508_CGO2-150x48.png 150w, https:\/\/www.tejwin.com\/wp-content\/uploads\/2508_CGO2-768x244.png 768w, https:\/\/www.tejwin.com\/wp-content\/uploads\/2508_CGO2.png 1320w\" sizes=\"(max-width: 1024px) 100vw, 1024px\" \/><figcaption class=\"wp-element-caption\">Holding period\uff1a 1 day\u3000\u3000\u3000Data period: January 2005\u2013June 2025)<\/figcaption><\/figure>\n\n\n\n<div style=\"height:25px\" aria-hidden=\"true\" class=\"wp-block-spacer\"><\/div>\n\n\n\n<h2 class=\"wp-block-heading\"><span class=\"ez-toc-section\" id=\"Risk_Factor_Model_Regression\"><\/span><a>Risk Factor Model Regression<\/a><span class=\"ez-toc-section-end\"><\/span><\/h2>\n\n\n\n<p>To test whether the returns of CGO-sorted portfolios can be explained by well-known systematic risk factors, we conduct time-series regressions of monthly excess returns against several classical asset pricing models. At the beginning of each month, stocks are grouped by their end-of-month CGO values, and portfolio returns are held for one month. These returns are then regressed on the factors of four benchmark models. The primary focus of this analysis is<strong> <mark style=\"background-color:rgba(0, 0, 0, 0)\" class=\"has-inline-color has-luminous-vivid-orange-color\">the intercept term (alpha)<\/mark><\/strong>. A significantly positive alpha indicates the presence of abnormal returns unexplained by the included risk factors. To account for serial correlation and heteroskedasticity in financial time series, all t-statistics are adjusted using the Newey\u2013West (1987) method.<\/p>\n\n\n\n<p>Table 5\uff1aCGO Portfolio Factor Model Regression Alpha<\/p>\n\n\n\n<figure class=\"wp-block-table is-style-stripes\"><table class=\"has-background has-fixed-layout\" style=\"background-color:#ffe9ae\"><thead><tr><th>Portfolio<\/th><th class=\"has-text-align-right\" data-align=\"right\">CAPM<\/th><th class=\"has-text-align-right\" data-align=\"right\">Fama-French <br>3 Factors<\/th><th class=\"has-text-align-right\" data-align=\"right\">Fama-French <br>5  Factors<\/th><th class=\"has-text-align-right\" data-align=\"right\">Fama-French <br>6 Factors<\/th><\/tr><\/thead><tbody><tr><td><strong>Bottom<br>Quantile<\/strong><br><strong>P1<\/strong><\/td><td class=\"has-text-align-right\" data-align=\"right\">0.0017<br>(0.58)<\/td><td class=\"has-text-align-right\" data-align=\"right\">-0.0015<br>(-0.76)<\/td><td class=\"has-text-align-right\" data-align=\"right\">0.0001<br>(0.03)<\/td><td class=\"has-text-align-right\" data-align=\"right\">0.0034<br>(1.62)<\/td><\/tr><tr><td><strong>Top<br>Quantile<\/strong><br><strong>P10<\/strong><\/td><td class=\"has-text-align-right\" data-align=\"right\"><strong>0.0106***<\/strong><br>(5.49)<\/td><td class=\"has-text-align-right\" data-align=\"right\"><strong>0.0082***<\/strong><br>(5.69)<\/td><td class=\"has-text-align-right\" data-align=\"right\"><strong>0.0090***<\/strong><br>(5.85)<\/td><td class=\"has-text-align-right\" data-align=\"right\"><strong>0.0077***<\/strong><br>(5.26)<\/td><\/tr><tr><td><strong>Spread<br>P10-P1<\/strong><\/td><td class=\"has-text-align-right\" data-align=\"right\"><strong>0.0088***<\/strong><br>(2.81)<\/td><td class=\"has-text-align-right\" data-align=\"right\"><strong>0.0097***<\/strong><br>(3.19)<\/td><td class=\"has-text-align-right\" data-align=\"right\"><strong>0.0090***<\/strong><br>(2.70)<\/td><td class=\"has-text-align-right\" data-align=\"right\">0.0043<br>(1.38)<\/td><\/tr><\/tbody><\/table><figcaption class=\"wp-element-caption\"><em>The values \u200b\u200bin parentheses are Newey-West adjusted t-test statistics. Significance level: * p&lt;0.01, ** p&lt;0.05, *** p&lt;0.1<\/em><\/figcaption><\/figure>\n\n\n\n<div style=\"height:25px\" aria-hidden=\"true\" class=\"wp-block-spacer\"><\/div>\n\n\n\n<p>Table 5 reports the regression alphas for the lowest-CGO portfolio (P1), the highest-CGO portfolio (P10), and the long\u2013short spread (P10\u2013P1).Key findings are as follows:<\/p>\n\n\n\n<ul class=\"wp-block-list\">\n<li><mark style=\"background-color:rgba(0, 0, 0, 0)\" class=\"has-inline-color has-luminous-vivid-orange-color\"><strong>High-CGO portfolios show significant alpha.<\/strong><br><\/mark>Under CAPM and Fama\u2013French models (FF3, FF5), the highest-CGO portfolio (P10) delivers strong and statistically significant alpha, indicating that CGO captures excess returns not explained by standard factors, mainly from the long side.<br><\/li>\n\n\n\n<li><mark style=\"background-color:rgba(0, 0, 0, 0)\" class=\"has-inline-color has-luminous-vivid-orange-color\"><strong>CGO overlaps with momentum<\/strong>.<br><\/mark>The long\u2013short spread is significant under CAPM, FF3, and FF5, but loses significance in the FF6 model once momentum is included, consistent with Grinblatt and Han (2005) that momentum largely reflects the Disposition Effect.<\/li>\n<\/ul>\n\n\n\n<h2 class=\"wp-block-heading\"><span class=\"ez-toc-section\" id=\"Information_Coefficient_IC_Analysis\"><\/span>Information Coefficient (IC) Analysis<span class=\"ez-toc-section-end\"><\/span><\/h2>\n\n\n\n<p>To further evaluate the predictive power of CGO, we calculate the Information Coefficient (IC)\u2014the Spearman rank correlation between CGO values and subsequent returns across different holding horizons.<\/p>\n\n\n\n<p>Table 6\uff1aSummary of Information Coefficients (ICs) for the CGO Factor at Different Holding Periods<\/p>\n\n\n\n<figure class=\"wp-block-table is-style-stripes\"><table class=\"has-background\" style=\"background-color:#ffe9ae\"><thead><tr><th><\/th><th class=\"has-text-align-right\" data-align=\"right\">1D<\/th><th class=\"has-text-align-right\" data-align=\"right\">5D<\/th><th class=\"has-text-align-right\" data-align=\"right\">10D<\/th><th class=\"has-text-align-right\" data-align=\"right\">20D<\/th><th class=\"has-text-align-right\" data-align=\"right\">40D<\/th><th class=\"has-text-align-right\" data-align=\"right\">60D<\/th><th class=\"has-text-align-right\" data-align=\"right\">100D<\/th><\/tr><\/thead><tbody><tr><td>IC Mean<\/td><td class=\"has-text-align-right\" data-align=\"right\">-0.001<\/td><td class=\"has-text-align-right\" data-align=\"right\">0.0008<\/td><td class=\"has-text-align-right\" data-align=\"right\">0.0065<\/td><td class=\"has-text-align-right\" data-align=\"right\">0.0079<\/td><td class=\"has-text-align-right\" data-align=\"right\">0.0252<\/td><td class=\"has-text-align-right\" data-align=\"right\">0.0448<\/td><td class=\"has-text-align-right\" data-align=\"right\">0.0635<\/td><\/tr><tr><td>IC Std<\/td><td class=\"has-text-align-right\" data-align=\"right\">0.1226<\/td><td class=\"has-text-align-right\" data-align=\"right\">0.1369<\/td><td class=\"has-text-align-right\" data-align=\"right\">0.1389<\/td><td class=\"has-text-align-right\" data-align=\"right\">0.1436<\/td><td class=\"has-text-align-right\" data-align=\"right\">0.1415<\/td><td class=\"has-text-align-right\" data-align=\"right\">0.1318<\/td><td class=\"has-text-align-right\" data-align=\"right\">0.1144<\/td><\/tr><tr><td>Risk Adjusted IC<\/td><td class=\"has-text-align-right\" data-align=\"right\">-0.008<\/td><td class=\"has-text-align-right\" data-align=\"right\">0.0062<\/td><td class=\"has-text-align-right\" data-align=\"right\">0.0465<\/td><td class=\"has-text-align-right\" data-align=\"right\">0.0552<\/td><td class=\"has-text-align-right\" data-align=\"right\">0.1781<\/td><td class=\"has-text-align-right\" data-align=\"right\">0.3401<\/td><td class=\"has-text-align-right\" data-align=\"right\">0.5554<\/td><\/tr><tr><td>IC t-value<\/td><td class=\"has-text-align-right\" data-align=\"right\">-0.5649<\/td><td class=\"has-text-align-right\" data-align=\"right\">0.4359<\/td><td class=\"has-text-align-right\" data-align=\"right\">3.2661<\/td><td class=\"has-text-align-right\" data-align=\"right\">3.8777<\/td><td class=\"has-text-align-right\" data-align=\"right\">12.5163<\/td><td class=\"has-text-align-right\" data-align=\"right\">23.9026<\/td><td class=\"has-text-align-right\" data-align=\"right\">39.0314<\/td><\/tr><tr><td>IC p-value<\/td><td class=\"has-text-align-right\" data-align=\"right\">0.5721<\/td><td class=\"has-text-align-right\" data-align=\"right\">0.663<\/td><td class=\"has-text-align-right\" data-align=\"right\"><strong>0.0011<\/strong>***<\/td><td class=\"has-text-align-right\" data-align=\"right\"><strong>0.0001<\/strong>***<\/td><td class=\"has-text-align-right\" data-align=\"right\"><strong>0<\/strong>***<\/td><td class=\"has-text-align-right\" data-align=\"right\"><strong>0<\/strong>***<\/td><td class=\"has-text-align-right\" data-align=\"right\"><strong>0<\/strong>***<\/td><\/tr><tr><td>IC Skewness<\/td><td class=\"has-text-align-right\" data-align=\"right\">-0.1767<\/td><td class=\"has-text-align-right\" data-align=\"right\">-0.3965<\/td><td class=\"has-text-align-right\" data-align=\"right\">-0.4118<\/td><td class=\"has-text-align-right\" data-align=\"right\">-0.5668<\/td><td class=\"has-text-align-right\" data-align=\"right\">-0.9367<\/td><td class=\"has-text-align-right\" data-align=\"right\">-1.1376<\/td><td class=\"has-text-align-right\" data-align=\"right\">-1.1015<\/td><\/tr><tr><td>IC Kurtosis<\/td><td class=\"has-text-align-right\" data-align=\"right\">1.6359<\/td><td class=\"has-text-align-right\" data-align=\"right\">1.3835<\/td><td class=\"has-text-align-right\" data-align=\"right\">1.1328<\/td><td class=\"has-text-align-right\" data-align=\"right\">0.7843<\/td><td class=\"has-text-align-right\" data-align=\"right\">1.4874<\/td><td class=\"has-text-align-right\" data-align=\"right\">2.3776<\/td><td class=\"has-text-align-right\" data-align=\"right\">2.7987<\/td><\/tr><\/tbody><\/table><figcaption class=\"wp-element-caption\"><em>(Data period: January 2005\u2013June 2025, Significance Levels: *** p&lt;0.01, ** p&lt;0.05, * p&lt;0.1)<\/em><\/figcaption><\/figure>\n\n\n\n<div style=\"height:25px\" aria-hidden=\"true\" class=\"wp-block-spacer\"><\/div>\n\n\n\n<p>Table 6 reports the IC statistics. The results highlight two key insights:<\/p>\n\n\n\n<ol class=\"wp-block-list\">\n<li><strong><mark style=\"background-color:rgba(0, 0, 0, 0)\" class=\"has-inline-color has-luminous-vivid-orange-color\">Predictive strength grows with horizon.<br><\/mark><\/strong>For short horizons (e.g., 1\u20135 days), the IC is weak and insignificant. However, as the holding period extends, IC values turn positive and steadily increase. At the 100-day horizon, the mean IC reaches 0.0635 with strong statistical significance (t = 39.03).<br><\/li>\n\n\n\n<li><strong><mark style=\"background-color:rgba(0, 0, 0, 0)\" class=\"has-inline-color has-luminous-vivid-orange-color\">Strong risk-adjusted performance.<br><\/mark><\/strong>The risk-adjusted IC (IR) rises sharply with longer horizons, reaching 0.555 at 100 days. In practice, IR values above 0.5 are considered excellent, underscoring CGO\u2019s reliability as a medium- to long-term stock selection factor.<\/li>\n<\/ol>\n\n\n\n<p>In sum, the IC analysis provides robust evidence that CGO is a powerful predictive signal in Taiwan\u2019s equity market, particularly over 60\u2013100 day horizons, where its forecasting ability is both stable and economically meaningful.<\/p>\n\n\n\n<h2 class=\"wp-block-heading\"><span class=\"ez-toc-section\" id=\"Conclusion\"><\/span>Conclusion<span class=\"ez-toc-section-end\"><\/span><\/h2>\n\n\n\n<p>The empirical analysis confirms that <mark style=\"background-color:rgba(0, 0, 0, 0)\" class=\"has-inline-color has-luminous-vivid-orange-color\"><strong>Capital Gain Overhang (CGO) is a meaningful and robust factor in Taiwan\u2019s equity market<\/strong><\/mark>. Across portfolio-sorting tests, return analysis, risk factor regressions, and IC evaluation, CGO consistently shows predictive power, particularly over medium- to long-term horizons. High-CGO portfolios deliver significantly higher future returns, with evidence of positive alpha beyond standard Fama\u2013French factors. However, its overlap with momentum suggests that CGO captures behavioral effects closely tied to the Disposition Effect.<\/p>\n\n\n\n<p>Overall, these findings establish CGO as a reliable behavioral factor with clear economic intuition and strong statistical support. The next step is to translate these insights into <strong>Factor Strategy \u2013 CGO<\/strong>, where we design and backtest strategies to evaluate how CGO can enhance portfolio performance.\u300d<\/p>\n\n\n\n<p class=\"has-background\" style=\"background-color:#ffe9ae\"><strong><em>\ud83d\udc49\u00a0<a href=\"https:\/\/www.tejwin.com\/en\/insight\/factor-strategy-idiosyncratic-volatility\/\">Continue reading to explore how <\/a><a href=\"https:\/\/www.tejwin.com\/en\/insight\/factor-strategy-capital-gain-overhang-part-2\/\" data-type=\"link\" data-id=\"https:\/\/www.tejwin.com\/en\/insight\/factor-strategy-capital-gain-overhang-part-2\/\">CGO <\/a><a href=\"https:\/\/www.tejwin.com\/en\/insight\/factor-strategy-idiosyncratic-volatility\/\">can be transformed from a predictive factor into actionable investment strategies<\/a>.<\/em><\/strong><\/p>\n\n\n\n<figure class=\"wp-block-embed is-type-wp-embed is-provider-tej wp-block-embed-tej\"><div class=\"wp-block-embed__wrapper\">\n<blockquote class=\"wp-embedded-content\" data-secret=\"o7sU3oJyLy\"><a href=\"https:\/\/www.tejwin.com\/en\/news\/factor-library\/\">Factor Library \u2013 Taiwan&#8217;s Factor Dataset for Quantitative Investing<\/a><\/blockquote><iframe class=\"wp-embedded-content\" sandbox=\"allow-scripts\" security=\"restricted\" style=\"position: absolute; visibility: hidden;\" title=\"&#8220;Factor Library \u2013 Taiwan&#8217;s Factor Dataset for Quantitative Investing&#8221; &#8212; TEJ\" src=\"https:\/\/www.tejwin.com\/en\/news\/factor-library\/embed\/#?secret=37VWWRvA23#?secret=o7sU3oJyLy\" data-secret=\"o7sU3oJyLy\" width=\"600\" height=\"338\" frameborder=\"0\" marginwidth=\"0\" marginheight=\"0\" scrolling=\"no\"><\/iframe>\n<\/div><\/figure>\n","protected":false},"excerpt":{"rendered":"<p>The origins of the momentum anomaly have long been debated, with multiple competing explanations. Among them, one of the most influential behavioral interpretations attributes momentum to the Disposition Effect, a systematic bias in investor decision-making. This article focuses on the Capital Gain Overhang (CGO) factor, specifically designed to quantify this behavioral bias. Using the Taiwan equity market as a case study, we examine CGO\u2019s predictive power as a stock selection indicator and evaluate its practical value through empirical analysis.<\/p>\n","protected":false},"featured_media":39411,"template":"","tags":[],"insight-category":[690,3509],"class_list":["post-39409","insight","type-insight","status-publish","has-post-thumbnail","hentry","insight-category-data-analysis","insight-category-fintech-en"],"acf":[],"_links":{"self":[{"href":"https:\/\/www.tejwin.com\/en\/wp-json\/wp\/v2\/insight\/39409","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/www.tejwin.com\/en\/wp-json\/wp\/v2\/insight"}],"about":[{"href":"https:\/\/www.tejwin.com\/en\/wp-json\/wp\/v2\/types\/insight"}],"version-history":[{"count":11,"href":"https:\/\/www.tejwin.com\/en\/wp-json\/wp\/v2\/insight\/39409\/revisions"}],"predecessor-version":[{"id":39428,"href":"https:\/\/www.tejwin.com\/en\/wp-json\/wp\/v2\/insight\/39409\/revisions\/39428"}],"wp:featuredmedia":[{"embeddable":true,"href":"https:\/\/www.tejwin.com\/en\/wp-json\/wp\/v2\/media\/39411"}],"wp:attachment":[{"href":"https:\/\/www.tejwin.com\/en\/wp-json\/wp\/v2\/media?parent=39409"}],"wp:term":[{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/www.tejwin.com\/en\/wp-json\/wp\/v2\/tags?post=39409"},{"taxonomy":"insight-category","embeddable":true,"href":"https:\/\/www.tejwin.com\/en\/wp-json\/wp\/v2\/insight-category?post=39409"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}