Momentum behavior in mutual funds investing in the
U.S. and Japan
Thesis MSc. Finance
Abstract
I investigate to what extent mutual funds investing in U.S and Japanese stocks engage in momentum behavior. Both U.S and Japan focused funds gain significant momentum returns compared to a benchmark portfolio which does not rebalance its holdings. The quarterly excess returns of momentum investing using U.S stocks are 1,3%, which is significantly greater than the returns gained from buying Japanese shares, which provide an investor with excess returns of 0,5% over the benchmark portfolio. Furthermore, I find U.S focused funds can achieve significantly higher profits from momentum investing during crisis periods, unlike funds investing in Japanese stocks, where the higher return is offset by increased risk. I find evidence that momentum investing is positively related the performance of mutual funds.
Keywords: momentum investing, momentum effect, contrarian investing JEL classifications: G10, G14, G23
2
I. Introduction
In a search to gain some additional capital, individual investors are moving to institutional investors to invest their assets. Over the past decades institutions such as banks, pension funds and mutual funds have become increasingly more influential in financial markets. With a total of 18,6 trillion dollars in holdings in 2016, U.S. mutual fund assets exceeded the GDP of the United States. The growing participation of mutual funds in the financial market, combined with the lack of empirical research performed on them makes it ever more important to investigate the behavior of mutual fund managers. This is fundamental since mutual funds are both important enough in financial markets and distinct enough in their objectives and constraints to deserve focused attention (Kaminsky et al., 2001). It is beneficial to focus on mutual funds individually, since we can evaluate the behavior of fund managers and the underlying investors. Especially managers’ choices regarding portfolio composition is required to receive special thought.
The technique I am investigating makes use of the tendency of stocks to persist in their performance (Jegadeesh and Titman, 1993). This implies securities that have performed well in the past tend to perform well in the future. Similarly, poor performing securities tend to perform poorly in the future. This finding is a violation of the efficient market hypothesis (EMH), which states that all new information is directly reported in stock prices (Fama, 1965). Price persistence behavior in stocks can be exploited by a so called momentum strategy, which involves buying stocks with good results in the past and shorting those that performed poorly.
3
investigated. This forms a gap in the literature of momentum investing which I strive to fill. The research questions I want to answer in this paper are:
- To what extent do mutual funds investing in the U.S. and Japan engage in momentum trading?
- Does this strategy change during crisis periods?
- Is there a positive relationship between momentum behavior and fund performance?
I investigate whether there are differences in the implication of the momentum trading strategy between mutual funds with a focus on U.S stocks and funds investing in Japanese companies. Furthermore, I investigate whether the implementation of this technique changes in times of crisis. In times of recession the stock prices are highly volatile, driving interest to investigate the consequential managerial response. The main question I want to answer is in what magnitude mutual fund managers commit to using a standardized trading technique called momentum trading. I contribute to earlier work by investigating momentum behavior during crisis periods.
The available quarterly portfolio data allows me to track the development of their actual portfolios. Unlike U.S funds, mutual funds originating from Japan are not obligated to enclose their quarterly holdings with the public, making it impossible to get quarterly data from Japanese mutual funds. In order to draw a comparison between the countries, I use funds that are geographically focused on the U.S. and Japan, rather than funds that originate from those countries. I selected funds that invest in companies originating in the U.S. and Japan. I am looking into the assets of 50 U.S. focused mutual equity funds, as well as 50 Japan focused mutual funds. I find the data on the quarterly composition of their portfolios. This data is updated in March, June, September and December. The timeframe for my sample is December 2006 to September 2019.
4
Grinblatt et al.(1995). This measure calculated how much a portfolio manager exposes his firm to previous winning stocks rather than losing assets.
With the 'winners over losers' method I detect that both the U.S and the Japan focused funds engage in momentum behavior. The GWT measure shows that for U.S. focused funds the excess returns gained in buying winners are 2,6 times higher than for Japanese funds (GWT U.S = 0,013 and GWT Japan = 0,005). In crisis periods the returns from buying winning stocks are significantly higher for the U.S, in contrast to funds investing in Japanese stocks, which do not gain significantly higher returns during crisis periods. Similar to Grinblatt et al. (1995) I find a positive relationship between momentum investing and mutual fund performance.
The remainder of paper is organized as follows: In section 2 I discuss the relevant literature regarding momentum trading. Section 3 shows the data and explains the methodology in depth. Section 4 will display my results and section 5 concludes my thesis.
II. Literature review
2.1 Efficient market theory
Several decades ago, the efficient market hypothesis (EMH) was widely accepted by academics. As documented by Fama (1965), it was believed that markets are efficient in processing information. This implies that all new information is immediately reflected in stock prices. It is often associated with the idea of random walk, since the information that arrives today will only affect today’s stock price. This is irrespective of yesterday's news, since that is already incorporated in the pricing of yesterday.
According to the EMH, technical analysis, which is the investigation of past stock prices to predict future prices, would yield no excess returns. Fundamental analysis, the study of company specific financial information such as earnings would not benefit an investor more than investing in a portfolio of selected stocks, accounting for the same risk.
5
according to the weak efficient market hypothesis, in contrast to technical analysis. According to the EMH the latter will not yield excess profit, since there is no new information about the stock that can be revealed by the analysis of past stock price.
Ever since the EMH is formulated scholars have tried to ‘beat the market’. The evidence of momentum strategies proves that there are opportunities to do so. There are two anomalies that work against the EMH: the overreaction hypothesis and price persistence theory. These theories originate from the field of behavioral finance. The overreaction hypothesis describes the tendency of traders to overreact to new information (de Bondt and Thaler, 1985). Price persistence theory refers to a tendency for security prices to continue their price pattern. Jegadeesh and Titman (1993) found that stocks that have performed relatively well (poorly) over the past three to six months continue to do well (poorly) over the next three to six months. Lo, Mamaysky and Wang (2000) also find that some of the stock price signals used by “technical analysts,” such as “head and shoulders” formations and “double bottoms,” may actually have predictive power. Price persistence is explained by assuming the investor is irrational (Barberis et al. 1998; Daniel et al. 1998). Alternatively, momentum profits could be explained as compensation for some unspecified fundamental risk (Fama,1998). This anomaly in the EMH gives rise to momentum strategies which involve buying winners and selling losers.
Effective momentum trading is a clear violation of the EMH. Many researchers have expressed their doubt regarding the effectiveness of the strategy. Black (1993) and MacKinlay (1995) rationalized the results by accusing scholars of ‘data mining’. Conrad and Kaul (1998) state the excess return of momentum investing can be explained as compensation for bearing increased risk or doubting the results, because of ‘statistical chance’ (Fama, 1998). Nevertheless, the profitability of momentum investing is confirmed in various studies testing many different markets (Rouwenhorst, 1998; Griffin, Ji and Martin, 2003; Doukas and McKnight, 2005; Gupta, Locke and Scrimgeour, 2013).
2.2 Momentum trading
6
portfolio. These portfolios are equally weighted and held for K months (K being 3, 6, 9 or 12 months) in which they cannot be rebalanced.
Jegadeesh and Titman (1993) showed it is possible to beat the market using a strategy focused on buying winners and selling losers, earning a return of 1,49% a month with a strategy which forms a portfolio based on 12 months past stock returns and holds these stocks for 3 months, including a 1 week lag between the formation and the holding period. This gives rise to the thought that good performing stocks are likely to perform well in the future and vice versa.
The momentum effect has been investigated in depth (Moskowitz et al.,2012; Novy-Marx, 2012; and Li et al., 2018). It is also investigated with specific focus on mutual funds (Grinblatt et al., 1995; Wermers, 1999). Results in the research of momentum trading has been contradictory. Carhart (1997) shows that U.S mutual funds engage in such practices. This is confirmed by other studies by Nofsinger and Sias (1999), Titman (2001), and Mulvey and Kim (2008). The opposite results are evident in studies by Gompers and Metrick (2001), and de Haan and Kakes (2011) who find no evidence of momentum trading. Tse (2015) states the methods used by Jegadeesh and Titman (1993) were totally insignificant and concludes that between the late 90s and 2010 the buy and hold strategy would beat the momentum strategy.
Jegadeesh and Titman (1993) state that momentum returns are consistent with delayed reactions to firm-specific information. The extreme winner and the extreme loser of the sample have a greater beta than the average of the sample. The winning and the losing securities have a smaller than average market capitalization. Additionally, the losers’ market cap is smaller than the winners. The authors conclude momentum returns are not caused by systematic market risk or a size premium. The risk premium associated with momentum profits is caused by firm-specific risk factors.
7
and high volatility. This suggests a strategy that combines momentum and volatility outperforms a strategy that solely focuses on either momentum or volatility.
Novy-Marx (2012) forms a strategy of choosing winner stocks based on an intermediate timespan (12 to 7 months). This method gains superior returns as well as significant alphas when compared to the Fama-French four factor model. De Bondt et al (1985) find contradictory results, displaying past losers beating past winners with losers’ returns of 25% more than winners. This result was later confirmed by de Bondt et al. (1987). This may be explained by the overreaction of investors on a negative price shock. Carhart (1997) investigates mutual firms’ short term persistence and finds that using momentum trading in buying mutual funds yields a yearly return of 8%.
Most studies focus on the U.S., but there is also evidence from other countries. In a cross-country study Chui et al. (2010) finds momentum strategies are profitable, achieving even higher Sharpe ratios than U.S. studies. Rouwenhorst (1998) investigates 12 European countries finding a zero-cost momentum portfolio buying medium term winners and selling medium term losers earn approximately 1 percent per month. Longer holding periods tend to decrease average returns. After examining the market beta of the portfolios Rouwenhorst (1998) states it is unlikely that the excess returns are explained by market risk. Furthermore, he finds the firm size of the winner and loser securities are smaller than the average firm size of the sample. Interestingly, losers seem to be smaller than winners. Rouwenhorst (1998) forms size decile portfolios finding momentum is present amongst all sizes, but the effect is larger amongst smaller firms. The results from 12 European countries confirm the findings in the U.S. by Jegadeesh and Titman (1993).
Griffin et al. (2003) investigate a sample consisting of 40 different countries. They find positive price momentum effects in all six American countries, 10 out of 14 Asian countries and 14 out of 17 European countries. Hameed and Kusnadi (2002) find that the magnitude of the momentum effect is least visible in the Asian-Pacific market, where there are no abnormal returns. This finding is confirmed by Griffin et al. (2003). Fama and French (2012) find positive momentum in all countries in their sample except for Japan. These findings pave the way for the comparison between the U.S. and Japan in this paper.
8
Chan et al (1996) distinguish between earnings momentum and price momentum to predict future stock returns. Earnings momentum is based on underreaction by the public on earnings announcements, whereas price momentum is influenced by several information sources regarding past returns. They test three different variables of earnings surprise variables and find positive results for each of them. The reported results for price momentum are stronger than for the earnings variables, finding price momentum returns of 8,8% in six months and 15,4% in 12 months after composing the portfolio. Chan et al (1996) conclude that earnings surprises as well as price information have slight predictive power for future security prices. This is explained by market underreaction to these types of information.
In a 40-year study of the London Stock Exchange, Galariotis et al (2007) find that both momentum and contrarian strategies can yield high returns. However, the profitable contrarian strategy disappears when it is adjusted for risk factors in the Fama-French three factor model. An international study by Fama and French (2012) looks into the relationship between momentum returns and firm size. They conclude momentum returns are higher for larger firms.
Whilst there are many studies confirming the presence of momentum behavior in securities, proving the economic significance tends to be challenging. In order for the momentum strategy to work, frequent trading is necessary. Therefore, traders experience significant transaction costs. These costs are not fixed and researchers have not reached consensus over what the correct implementation should be. Jegadeesh and Titman (1993) and Lui et al. (1999) use an estimate of 0,5%, which is based on average transaction cost. However, scholars have shown this figure underestimates the observed costs (Lesmond, et al., 2004; Korajczyk and Sadka, 2004; Agyei-Ampomah, 2007). They find stock specific characteristics are the main driver of bid and ask spreads and therefore transaction costs should be based on firm characteristics. Korajczyk and Sadka (2004) find momentum strategies that showed to gain high excess returns still hold a positive return, when accounting for trading costs. However, the impact of trading costs causes momentum profits to significantly decline. Lesmond et al (2004) concludes that a momentum strategy requires frequent trading, which leads to high transaction costs. This diminishes the excess return gained from the strategy.
2.3 Reasons for momentum trading
9
(1990) believes traders are rational and will anticipate the price will rise, because of the presence of noise traders. Bhojraj and Swaminathan (2006) argue that transaction costs and data snooping biases can explain excess returns. However the robustness of the momentum effect is confirmed by several studies (Fama and French, 1996; Rouwenhorst, 1997).
Daniel et al. (1998) and Barberis et al. (1998) provide another explanation using behavioral investor psychology. In both papers a model is provided based on underreaction and overreaction on stock prices, finding positive private signals are often valued too highly by professional investors. These models are built on the EMH, which states all available information should be incorporated in today’s security price. This implies when a stock gains positive excess returns in the period after good news is released, the market has underreacted. Similarly, we speak of market underreaction when security prices drop after an announcing of bad news. In the next period security prices are slowly corrected for. This underreaction of the market is a short-run process, used by Daniel et al. (1998) and and Barberis et al. (1998) to explain 3 through 12-month momentum effects. In contrast, overreaction of the market is a long run process, which occurs after a consecutive pattern of positive (or negative) information. This causes investors to become overconfident (underconfident) and leads to security prices being overvalued (undervalued). Eventually this mispricing is corrected for, explaining price reversals in security prices in a three to five-year period.
The model of Daniel et al. (1998) is used by Moskovitz and Grinblatt (1999) to give a rationale for industry momentum. Since new and changing industries are difficult to value, they find investors tend to misprice these industries when they are employed there, due to overconfidence and self-attribution. These findings also hold for industries in which an investor has much knowledge. Hirshleifer (2001) states investors’ overconfidence increases in absence of accurate feedback of firm specific information. This means overconfidence should be stronger when a certain firm is harder to value. Zhang (2006) confirms this finding, arguing misevaluation is the strongest when uncertainty is high and information provided is poor. In the context of momentum investing this implies that investors tend to underreact to new information when its effect on firm performance is ambiguous.
10
investors. These categories are 'newswatchers' and 'momentum traders'. The different forms of bounded rationality between these two groups helps to explain momentum behavior in securities. ‘Newswatchers’ use private information regarding fundamentals. They are not interested in past prices. In contrast, ‘momentum traders’ base their decisions on price changes. According to Hong and Stein (1999) public information does not reflect directly in stock prices, and ‘newswatchers’ may fail to extract all necessary information from prices, creating opportunities for momentum trading.
Lee and Swaminathan (2000) argue that price momentum can be predicted by past trading volume. Their test is similar to that of Jegadeesh and Titman (1993), but they included trading volume analysis. Their find that high volume stocks exhibit higher momentum than their counterparts. Additionally, they found when controlled for momentum low volume stocks outperform high volume stocks.
There are examples of researchers linking momentum profits to firm rating. Avramov et al. (2007) find no significant momentum profits for stocks with a rating between BB and AAA, whereas lower rated securities do exhibit significant momentum profits. The effect of firm size is researched by Jegadeesh and Titman (1993) and Fama and French (2008). They find momentum is present for all stock sizes.
Chui et al. (2010) provide a link between momentum and the cultural dimensions of Hofstede (2001). The cultural dimension ‘individualism’ is correlated with overreaction in the market. This is caused by overconfidence and attribution bias. Chui et al (2010) find a positive relationship between ‘individualism’ and trading volume, volatility and magnitude of the momentum behavior in stocks.
11
Besides individual stock momentum, industry portfolios also exhibit significant momentum behavior (Moskowitz and Grinblatt, 1999). While the researchers do not find the rationale for the momentum behavior, they find two differences between individual stock momentum and industry momentum. Firstly, individual stock momentum is mainly driven by shorting losers, whereas industry momentum arises mainly from buying winners. Secondly industry momentum is strongest in the very short run (1 month holding period), where the momentum of individual stocks is stronger in a holding period of 3 through 12 months.
2.4 Mutual funds
For the past decades, mutual funds have grown significantly in size. As a result of this growth, mutual funds have become an influential player in the market. The interaction between mutual funds size and return is studied by Titman and Grinblatt (1989), Carhart (1997), Chen et al. (2004) and Ferreira et al. (2012). The results regarding this research are twofold. The argument can be made that larger funds can use their assets to access more private information and perform more research on individual firms, thereby making them less vulnerable for herding behavior and momentum strategies. However, larger mutual funds on average have a more complex portfolio, which makes researching individual firms costly and time-consuming. Titman and Grinblatt (1989) find smaller funds exhibit more excess returns compared to their counterparts, but small funds also face higher transaction costs. They conclude actual net returns are unrelated to mutual fund size. There is no consensus about the effect of the size of a mutual fund on its implementation of momentum strategies.
There have been extensive studies about the performances of mutual funds’ investors. In a study of the quarterly holdings of U.S. mutual funds, Grinblatt et al. (1995) find that 77% of the funds show momentum behavior and significantly outperform the benchmark. This is in line with Grinblatt and Titman (1993) who document about the abnormal investment performance of mutual funds. Karoui et al. (2009) find evidence that the performance of mutual funds declines after three years. This is explained by the higher exposure on small stocks (thus higher level of unsystematic risk) these funds take.
12
following momentum strategies and are mostly contrarian. This is in line with the conclusion by Gomper and Metrick (2001) who, when looking at changes in firm level institutional ownership, find no evidence of momentum strategies.
When analyzing the performance of mutual funds, special attention has to be taken regarding the ‘incubation bias’. This occurs when funds are kept private and only the outperforming stocks are displayed to the public (Evans, 2007). In later research Evans (2010) finds that 23% of funds were incubated and explains this is merely to attract additional capital and not to identify managerial performance. When not corrected for ‘incubation bias’ the performance results are shifted upwards. However, Evans (2010) finds this effect is not significant compared to non-incubated funds, so I will take no special control measures regarding this.
2.5 Momentum in times of crisis
In times of crisis momentum strategies are vulnerable. Daniel and Moskowitz (2013) find that the strong returns associated with this period are interrupted with strong reversals. They argue that in a normal economic climate the market experiences price momentum, because the market underreacts to changes in past pricing. However, in times of crisis a large premium is connected to past losers. After a period of strong negative returns, the portfolio consists of winners with a high beta and losers with a low beta, making the beta of the WML portfolio negative. When the market recuperates, the loser portfolio experiences high returns, leading to a momentum crash. This is confirmed by Grundy and Martin (2001). They find that momentum strategies have significant negative betas in worsening markets.
Barroso and Santa-Clara (2015) find that when market stress reaches high levels, momentum trading is doomed to fail. They show that these strategies lost 73.42% in three months during the 2009 crisis. Kaminsky et al. (2001) find evidence that mutual funds systematically buy winners and sell losers and that the momentum strategy is stronger during crisis periods. Specifically, contemporaneous trading (buying/selling current winners/losers) increases. Lagged momentum trading (buying/selling past winners/losers) does not change. Furthermore, they find that this effect is equally strong for managers and investors.
13
when firm size increases. Furthermore, momentum profits decline when book-to-market ratio increases. In a recent study on mutual and hedge funds Grinblatt et al. (2016) find that whilst hedge funds are contrarians, mutual funds engage in momentum investing. This momentum trading by mutual funds increased during crisis, between mid-2007 and mid-2008. Prior research indicates that momentum behavior might increase during crisis periods.
III. Data and methodology
3.1 Data
For my research I use the data of U.S. and Japan focused mutual funds. I will first calculate the ‘winners over losers’ and the ‘winners minus losers’ ratios to detect forms of momentum trading. I test for momentum trading using the GWT measure as portrayed by Grinblatt et al. (1995). For this research I am looking into the assets of 50 U.S. focused mutual equity funds, as well as 50 Japan focused mutual funds. A dataset is formed using open-end mutual equity growth funds. For U.S. and Japan, the 50 funds with the highest value of net assets are included. Mutual funds who only reported yearly portfolio holding information are excluded.
14
Table 1. Descriptive statistics of U.S. and Japanese focused mutual funds
This table reports the total net assets (TNA), net asset value (NAV), which is computed by subtracting a funds’ liabilities from fund value and dividing by the number of shares outstanding; expense ratio, computed by dividing the operating expenses over net assets; 10-year performance growth and 5 year Sharpe ratio, which measures the performance of a fund compared to a risk free asset whilst accounting for risk. The sample size is 50 U.S focused and 50 Japan focused funds. The timespan is Dec. 2006 through Sep. 2019.
Mean Median Std. Deviation
U.S. Japan U.S. Japan U.S. Japan
TNA (Million $) 10.054 621 4.108 325 18.428 816 NAV($) 53,54 51,51 44,89 15,34 33,97 84,39 Expense ratio 0,77% 1,39% 0,79% 1,53% 0,20% 0,46% 10 years growth 286,37% 117,22% 282,95% 90,77% 37,12% 81,13% 5 years Sharpe 0,24 0,19 0,22 0,19 0,04 0,07 3.2 Survivorship bias
Survivorship bias originates from the exclusion of funds that did not exist during the complete timeframe of the research (Dec. 2006 - Sep. 2019). I conduct my research with a dataset subject to survivorship bias, similar to Grinblatt and Titman (1992). They explained this choice stating that investors are only interested in stocks they can actually invest in. Furthermore, controlling for survivorship bias implies consistently bad performing firms are eliminated due to bankruptcy, whereas firms with multiple performance reversals are included. A dataset constructed in this manner will give the false impression that performance reversals occur very frequently. Therefore, the dataset will likely show no presence of momentum behavior. This argument is confirmed by Bilson et al. (2005) who find that the survivorship bias does not affect their results.
15
3.3 Detecting momentum
Jegadeesh and Titman (1993) describe an intuitive strategy to detect institutional momentum. Hereby the stocks of the S&P 500 and Nikkei 225 indices are ranked in ascending order based on their three month previous returns. Then all the securities are split in ten deciles, where the top decile is the “losers” decile and the bottom decile is called the “winners” decile. For each quarter the following ratio is calculated:
(eq. 3.1) 𝑀",$ = '"(()*+
,,-./+)*+
,,-Here Winners indicates the sum of the weights of stocks in the top decile which a certain mutual fund i possesses at time t. Additionally, Losers indicates the sum of weights of stocks in the bottom decile for fund i at time t. Whenever 𝑀",$ > 1, a firm has exposed itself more to winning stocks, or less to losing stocks, In both cases we speak of a momentum strategy. Similarly, a fund going long in losing stocks or shorting winning stocks is implementing a contrarian strategy. This is detected when 𝑀",$ < 1. Equation 3.1 is used to test whether U.S. and Japan focused funds engage in momentum behavior. Additionally, I test the conclusion that momentum behavior is less present in Japan (Griffin et al. 2003; Hameed and Kusnadi, 2002 and Fama and French, 2012). This leads to the following hypotheses:
𝐻4,5 : The winners over losers ratio is on average equal to 1 𝐻4,4: The winners over losers ratio is on average greater than 1
𝐻6,5 : The winners over losers ratio is the same for U.S and Japan focused mutual funds 𝐻6,4: The winners over losers ratio is greater for funds focused on the U.S compared to their Japan focused counterparts
The main complication of the winners over losers ratio, is that it cannot be calculated when “losers” equals zero. Therefore, I also calculate the difference between the weights:
16
The interpretation of this measure is virtually similar to the winners over losers measure. A momentum strategy is detected when 𝑀",$ > 0 and contrarian behavior is detected when 𝑀",$ < 0. The following null and alternative hypotheses are tested with equation 3.2:
𝐻A,5 : The winners minus losers ratio is on average equal to zero 𝐻A,4: The winners minus losers ratio is on average greater than zero
𝐻B,5 : The winners minus losers ratio is the same for U.S and Japan focused mutual funds 𝐻B,4: The winners minus losers ratio is significantly greater for funds focused on the U.S compared to their Japan focused counterparts
3.4 Measuring momentum
I use the method as used by Grinblatt et al. (1995) called GWT method to measure the extent to which mutual funds focused on U.S. and Japan engage in momentum investing. This statistic is used in many other studies (Badrinath and Wahal 2002; De Haan and Kakes 2011; Curcuru et al. 2011). This measure shows how much a manager constructs his portfolio towards winners and stays away from losers. It shows the difference between a mutual fund which would have rebalanced the portfolio at the beginning of the quarter, and a fund that kept its portfolio the same during this time period. For each fund the changes in its portfolio are monitored. Each change in holdings of a certain stock is multiplied by the quarterly return of that stock. Aggregating all changes over the timespan of 13 years gives the GWT measure.
Mathematically:
(eq. 3.3) 𝐺𝑊𝑇 = J (𝑤G,$− 𝑤G,$H4)
GK4 L
$K4 𝑅G,$HN
17
straightforward to investigate whether momentum behavior diminishes during crisis periods. Equation 3.3 is used to test the following hypotheses:
𝐻O,5 : The GWT ratio is on average equal to zero 𝐻O,4: The GWT ratio is on average greater than zero
𝐻P,5 : The GWT ratio is the same for U.S and Japan focused mutual funds
𝐻P,4: The GWT ratio is greater for funds focused on the U.S compared to their Japan focused counterparts
The GWT ratio is used to test whether mutual funds increase momentum investing during crisis periods, as stated by Chui et al (2000), Grinblatt et al. (2016) and Kaminsky et al. (2001). The null and alternative hypotheses are:
𝐻Q,5 : The GWT ratio is the same during crisis periods compared to non-crisis periods. 𝐻Q,4: The GWT ratio in crisis periods is greater compared to non-crisis periods.
I test the findings of Grinblatt et al. (1995), stating momentum investing is positively related to mutual funds’ performance. For this test the GWT measure in equation 3.3 is used, leading the following hypotheses:
𝐻R,5 : There is no positive relationship between GWT and mutual fund performance. 𝐻R,4: There is a positive relationship between GWT and mutual fund performance.
IV. Empirical results
18
4.1. Detecting momentum with ‘winners’ and ‘losers’
Each quarter, the winners and losers are computed, where winners are the 10% best performing stocks in the S&P 500 (U.S.) or Nikkei 225 (Japan), and losers the 10% worst performers. These quarterly results are aggregated to provide a firm specific winners ratio and losers ratio. The W/L and W-L ratios are computed using the aggregated results and displayed in table 2. More detailed results are present in the Appendix. The W/L ratio should be equal to 1 if no momentum investing would have been used. Values greater than 1 indicate managers have been buying winning stocks selling losers. Table 2 shows both U.S and Japan focused funds exhibit momentum behavior.
Funds investing in Japan (where mean W/L is 1,617) are on average 61,7 % more exposed to winning stocks compared to losers. In addition, I find a significant difference between funds investing in the U.S. and those investing in Japan, confirming earlier research showing momentum investing is less present in Asian countries (Griffin et al. 2003; Hameed and Kusnadi, 2002 and Fama and French, 2012).
Although the W/L measure is intuitive and easy to interpret, there are some limitations connected to using the method. The main problem is that is cannot be used when the holding of losing stocks equals zero, leading to valuable data being left out. Furthermore, extreme results are possible while using this ratio (Maximum W/L for U.S. equals 26,788). This inflates the mean of the sample, and weakens its relevance.
19
Table 2
This table reports the results from the ‘winners over losers’ and ‘winners minus losers’ methods of detecting momentum. Quarterly data over 13 years is used, resulting in 52 observations. The first two columns show the coefficients for the ‘winners over losers’ method for 50 U.S and 50 Japan focused funds. The third column shows the difference between the countries. The results for the ‘winners minus losers’ method are displayed in the same manner. P-values for corresponding t-tests are reported in parenthesis. Significance levels of 1% are indicated with *.
W/L W-L
U.S. Japan Difference
U.S. Japan
U.S. Japan Difference
U.S. Japan Mean 4,700* (0,000) 1,617* (0,000) 3,083* (0,000) 0,034* (0,000) 0,012* (0,000) 0,022* (0,000) Median 3,670 1,469 0,034 0,011 Std. Dev. 3,965 0,971 0,017 0,009 Maximum 26,788 4,269 0,090 0,030 Minimum 1,157 0,006 -0,003 -0,020
4.2 Measuring momentum with the GWT measure
The GWT measure is calculated by multiplying the changes in holdings of a security by its quarterly return for the complete time period (Dec. 2006 – Sep. 2019). These results are aggregated and time series means are calculated. These are displayed in table 3. Detailed results are found in the Appendix. This measure can be interpreted as the increased return a fund observes when it adjusts its portfolio to last quarters winning stocks compared to a fund that did not adjust its holdings. The GWT measures are significant at a 1% level for all funds, rejecting the null hypothesis of no momentum investing during this time span. These results confirm the findings of Grinblatt et al. (1995) who showed presence of momentum in mutual funds.
20
To interpret this finding consider a benchmark fund which does not rebalance its portfolio and has $100 to invest. Implementing the momentum strategy on U.S firms will provide this firm with an average quarterly return of 1,3%, resulting in $101,30. Investing in Japanese stocks less profitable, since it merely provides an average return of 0,5%, resulting in $100,50. This confirms the findings displayed by Griffin et al. (2003), who indicate that momentum is less present in Asian countries.
I have distinguished between momentum funds and contrarian funds in order to provide a straightforward overview of the scope of momentum investing. It can be seen that 95 of the 100 mutual funds (50 U.S. focused and 45 Japan focused) gain positive returns from exhibiting momentum behavior.
Table 3.
This table reports the results from GWT method. Similar to table 2, Quarterly data over 13 years is used, resulting in 51 observations, since the first observation drops out. The first two columns show the coefficients for GWT method for 50 U.S and 50 Japan focused funds, indicated by the number in parentheses. The third column shows the difference between the countries. The fourth and fifth column report the GWT coefficients for all funds engaging in momentum behavior. Results for contrarian funds are reported in the last two columns. P-values for corresponding t-tests are reported in parenthesis. Significance levels of 1% are indicated with *.
GWT All funds Momentum funds Contrarian
funds
U.S. (50) Japan(50) Difference
U.S.Japan
U.S. (50) Japan(45) U.S.(0) Japan (5)
Mean 0,013* (0,000) (0,000) 0,005* (0,000) 0,008* (0,000) 0,013* (0,000) 0,006* - (0,001) 0,002* Median 0,012 0,006 0,012 0,006 - -0,001 Std. Dev. 0,006 0,005 0,006 0,004 - 0,004 Maximum 0,037 0,018 0,037 0,018 - -0,000 Minimum 0,001 -0,009 0,001 0,000 - -0,009
4.3 Momentum investing in times of crisis
21
Wyplosz 2009). This is because the economy in Japan was primarily harmed by the increase in energy and other commodity prices. This shock started in the second quarter of 2008, therefore the timespan for ‘in crisis’ is March 2008 through June 2009. The ‘not in crisis ‘period for both countries is September 2009 through September 2019.
The results are displayed in table 4. For both countries, the GWT measure is positive and significantly larger when a recession occurs. A U.S. firm will earn 2.2% higher returns when it adjusts its portfolio to winners in times of crisis, which is significantly different from the 1.1% it would earn when there is no recession. In contrast, I find no significant evidence for the hypothesis that funds implementing momentum investing in Japanese stocks achieve higher returns in crisis periods. This can be explained by stocks being more volatile in crisis periods, therefore the variance in the GWT increases accordingly. This is illustrated by ‘in crisis’ for Japan, where GWT ranges from -5,6% to 7,4% when in a recession compared to 0,4% to 1,8% when not in crisis. This disparity of GWT in crisis periods invalidates statements that momentum investing is more profitable during these periods, since the higher mean GWT that a fund gains is offset by significantly higher risk.
Table 4.
This table reports the GWT coefficients for 50 U.S and 50 Japan focused mutual funds. The first two columns displays U.S mutual funds for crisis and non-crisis periods. The third column shows the difference between these periods. The timespan for crisis for the U.S. is Dec. 2007 – Jun. 2009. Similarly, fourth and the fifth column display the GWT coefficients for Japan focused funds, where the crisis timespan is Mar. 2008 – Jun. 2009. The difference between these periods shown in the last column. The timespan for non-crisis period is Sep. 2009 – Sep. 2019 for both countries. P-values for corresponding t-tests are reported in parenthesis. Significance levels of 1% are indicated with *.
GWT U.S. Japan
In crisis No crisis Difference In crisis No crisis Difference
22
4.4 Correlation between Momentum investing and fund performance
The correlation between momentum, indicated by the GWT measure and fund performance is displayed in table 5. My findings are in line with those of Grinblatt et al. (1995) showing a positive relationship between momentum behavior and fund performance. I find the statistically significant correlation coefficients of 0,068 for U.S. and 0,268 for Japan focused funds. I have split the sample in funds with a positive and negative correlation, showing for 80 out of 100 funds (34 out of 50 U.S. focused and 46 out of 50 Japanese focused) momentum investing is positively correlated with its performance.
V. Conclusion
5.1 Contributions
In this paper I research the topic 'smart money', which investigates whether fund managers have superior timing or apply certain techniques to gain profits in the stock market. I specifically focus on the technique of momentum investing, where past winners are bought and past losers are sold. This build on research by Grinblatt et al. (1995). I contribute to this paper in three ways. Firstly, a 25 years old investigation is executed using modern time data from 2006-2019. Secondly, the inclusion of Japan focused funds allows for a comparison between market. Thirdly, I investigate mutual fund behavior in crisis periods.
Table 5.
This table displays the correlation between GWT measure and firm performance. The first two columns include all 50 U.S. and Japan focused mutual funds, indicated by the number in parentheses. The third and fourth column reports the coefficient for all firms with a positive correlation, whereas the last two columns show the values for funds where GWT and performance are negatively related. P-values for corresponding t-tests are reported in parenthesis. Significance levels of 1% are indicated with *.
All Funds Positive correlation Negative correlation
U.S.(50) Japan(50) U.S.(34) Japan(46) U.S.(16) Japan(4)
23
In this research I adapt the technique defined by Jegadeesh and Titman (1993) in order to detect momentum behavior. With this method stocks are ranked by their returns, split up into deciles and presence of momentum can be shown by comparing the best decile with the worst. Momentum is measured similarly as performed by Grinblatt et al (1995). I calculate the GWT measure, which shows how much returns a fund would gain when it rebalances its portfolio towards last periods winners compared to a non-rebalancing fund.
Both measures give statistically significant results, indicating both funds investing in U.S. and Japanese stocks engage in momentum behavior. The positive GWT coefficient implies that this rebalancing is yielding excess returns for these funds.
I investigate the momentum strategy in crisis periods by calculating the GWT measure during the timespan of the financial crisis. A U.S. firm will earn 2.2% higher returns when it adjusts its portfolio to winners in times of crisis, which is significantly different from the 1.1% it would earn when there is no recession. This does not hold for funds investing in Japanese stocks, where there is no significant difference between crisis and non-crisis periods. Unsurprisingly, when in a recession the technique of picking past winners proves to be much riskier than in non-crisis times. Lastly I show there is a positive correlation between momentum investing and mutual fund performance, confirming the finding of Grinblatt et al (1995).
5.2 Limitations regarding the 'smart money' puzzle
Whilst it is tempting to draw conclusions regarding momentum trading, we have to be cautious to interpret the results regarding this subject. This is because the mere presence of momentum trading alone does not necessarily imply managers are following this trading strategy. Naturally, it could be the case that managers base their decisions on certain fund characteristics and patterns. This explains why sophisticated investors are trying to exploit the momentum effect in funds, in which case we observe ‘smart money’ investing. However, another possibility is that managers naively chase the stocks which were recent winners, making them incidentally benefit from the momentum effect. It is crucial to distinguish between those two motives, because the former option provides a possible explanation for the growth in actively managed mutual funds, while the latter does not.
5.3 Limitations regarding my research
24
S&P 500 and Nikkei 225 index. Although most mutual funds in my dataset only hold stocks from this index, inclusion of the NASDAQ, MSCI, NYSE, and AMEX would improve the scope of funds in the research.
Two methodological issues arise from using the GWT measure. The first issue arises because the portfolio holdings are presented at the end of the quarter, however firms rebalance their holding constantly. This implies rebalancing a portfolio could have been done 1 or 2 months prior, which gives a timing mismatch when a compare the holdings to the stock indexes.
As pointed out by Badrinath and Wahal (2002) the GWT equation does not account for the entry and exit of new stocks in a portfolio. This accordingly affects the weight of the other stocks. However, Elton et al. (2010) showed that GWT values do not change significantly when accounting for entry and exit of stocks.
25
Appendix
Testing the mean of two samples
Assume two samples: 𝑥4L and 𝑥6L ,where 𝑇: 1,2 … . . , 𝑁
Testing the equality of mean of those samples is done by a T-test. We test: 𝐻5: 𝑥4 = 𝑥6 versus 𝐻4: 𝑥4 ≠ 𝑥6
The test statistic for this is: 𝑡 =[\H [] ^] _` ^] _ , assuming normality N(0,1)
Descriptive statistics regarding the 50 U.S. and 50 Japan focused mutual funds
Mean Median Std Min Max Q1 Q3
26
The W/L and W-L statistics for U.S. focused mutual funds
Winners/Losers Winners-Losers
Fund Name Mean Median Std. dev Mean Median Std. dev
27 T Rowe Price New America Growth Fund 2,522 1,671 3,617 0,020 0,027 0,062 Sequoia Fund 1,157 0,000 3,544 0,007 0,000 0,059 PGIM Jennison Growth Fund 3,692 1,754 5,430 0,049 0,066 0,071 American Century Select Fund 4,221 1,149 14,445 0,027 0,026 0,066 Fidelity Stock Fund 26,788 0,627 142,034 0,090 0,092 0,074 Nicholas Fund 2,089 0,340 4,663 0,010 0,000 0,050 Elfun Trusts 2,083 1,131 3,019 0,007 0,012 0,055 Franklin Growth Opportunities Fund 4,459 1,837 7,843 0,045 0,038 0,069 Morgan StanleyPortfolio 2,890 1,172 4,633 0,057 0,056 0,096 Principal LargeCap Growth Fund 3,167 1,834 4,501 0,030 0,031 0,060 ClearBridge Large Cap Growth Fund 1,758 1,253 1,878 0,016 0,026 0,074 Delaware US Growth Fund 1,844 0,764 2,907 0,026 0,036 0,090 Laudus US Large Cap Growth Fund 4,245 0,921 11,860 0,039 0,026 0,077
Ivy Large Cap
28
The W/L and W-L statistics for Japan focused mutual funds
Winners/Losers Winners-Losers
Fund Name Mean Median Std. dev Mean Median Std. dev
Baillie Gifford Japanese 1,771 0,922 3,945 0,017 0,009 0,046 JPM Japan Equity 0,695 0,000 1,543 0,019 0,000 0,040 Man GLG Japan CoreAlpha Retail 1,034 0,226 2,087 -0,020 -0,023 0,073 Schroder ISF Japanese Equity 2,161 0,896 2,774 0,019 0,020 0,041
GAM Star Japan
Leaders USD 0,821 0,137 1,513 0,005 0,002 0,044 Matthews Japan Fund 1,063 0,000 1,989 0,015 0,004 0,037 Threadneedle Japan Retail 2,357 1,099 3,478 0,020 0,015 0,048 Schroder ISF Japanese Opportunities 1,385 0,290 2,528 0,009 0,004 0,033 T Rowe Price Japan Fund 1,741 0,569 3,294 0,015 0,003 0,041 LF Morant Wright Japan 0,215 0,000 0,599 0,006 0,000 0,023 Hennessy Japan Fund 1,126 0,000 3,484 0,019 0,000 0,053 Aviva Investors Japon 1,214 0,709 1,722 -0,002 0,000 0,041 Handelsbanken Japan Tema 1,698 1,386 1,884 0,026 0,022 0,042
GAM Star Japan
29
Interfund Equity
Japan 1,450 1,041 1,520 0,004 0,002 0,038
Standard Life Inv
Japan 3,253 2,126 3,940 0,029 0,039 0,051 AXA Framlington Japan 1,138 0,921 1,509 0,007 0,006 0,036 Scottish Widows Japan Growth 1,598 1,176 2,776 0,003 0,008 0,045 GAM Multistock - Japan 1,628 0,925 1,820 0,008 0,003 0,042 Lansforsakringar Japanfond 1,287 1,065 1,237 0,009 0,005 0,032 Fidelity Funds - Japan 3,335 0,843 8,337 0,021 0,020 0,048 Fonditalia Equity Japan 1,425 0,984 1,490 0,004 0,001 0,040 Comgest Growth Japan 4,269 0,060 12,969 0,012 0,003 0,047 BGF Japan Small & MidCap Opportunities 1,718 0,961 2,133 0,018 0,014 0,025 AS SICAV I - Japanese Equity 0,371 0,000 0,815 0,010 0,000 0,049 T Rowe Price Japanese Equity 1,538 0,243 2,751 0,017 0,009 0,040 Candriam Equities L Japan 1,488 1,074 1,724 0,009 0,004 0,038
SLI Glo SICAV
Japanese Equities 3,587 1,339 7,247 0,027 0,021 0,045 Goldman Sachs Japan Equity 2,798 1,244 4,358 0,018 0,011 0,035 JPMorgan Japan Equity 1,647 0,829 2,587 0,025 0,020 0,044 BCV Japac 1,379 0,000 4,201 0,008 0,000 0,036 BNP Paribas Japan Equity 1,849 1,034 2,713 0,003 0,002 0,047 SEB Japanfond 2,514 1,161 4,022 0,005 0,008 0,030 Fidelity Funds - Japan Advantage 3,272 1,663 5,218 0,026 0,032 0,040 Legg Mason IF Japan Equity 0,006 0,000 0,043 0,020 0,000 0,045
Fidelity Inst Japan 2,706 0,296 5,037 0,023 0,001 0,042
30
The GWT statistics for U.S. focused mutual funds
GWT (full time period) GWT (In crisis) GWT (not in crisis)
Fund Name Mean Median Std. dev Mean Median Mean Median
31 T Rowe Price New America Growth Fund 0,006 0,005 0,007 0,003 0,002 0,006 0,005 Sequoia Fund 0,001 0,003 0,018 0,001 0,004 0,000 0,002 PGIM Jennison Growth Fund 0,015 0,017 0,020 0,027 0,028 0,015 0,015 American Century Select Fund 0,010 0,008 0,007 0,014 0,012 0,009 0,008 Fidelity Stock Fund 0,037 0,035 0,037 0,074 0,054 0,032 0,033 Nicholas Fund 0,006 0,006 0,005 0,010 0,008 0,006 0,005 Elfun Trusts 0,007 0,006 0,005 0,015 0,013 0,006 0,005 Franklin Growth Opportunities Fund 0,012 0,010 0,011 0,012 0,009 0,013 0,010 Morgan StanleyPortfolio 0,012 0,011 0,013 0,011 0,011 0,011 0,011 Principal LargeCap Growth Fund 0,011 0,010 0,007 0,018 0,018 0,009 0,009 ClearBridge Large Cap Growth Fund 0,007 0,006 0,007 0,016 0,011 0,005 0,005 Delaware US Growth Fund 0,009 0,008 0,009 0,012 0,021 0,009 0,008 Laudus US Large Cap Growth Fund 0,012 0,011 0,033 -0,002 0,016 0,015 0,011
Ivy Large Cap
32
The GWT statistics for Japan focused mutual funds
GWT (full time period) GWT (in crisis) GWT (not in crisis)
Fund Name Mean Median Std. dev Mean Median Mean Median
Baillie Gifford Japanese 0,002 0,005 0,024 0,011 0,010 0,000 0,005 JPM Japan Equity 0,010 - 0,037 - - 0,012 0,005 Man GLG Japan CoreAlpha Retail -0,009 -0,004 0,027 -0,044 -0,023 -0,005 -0,004 Schroder ISF Japanese Equity 0,010 0,005 0,035 0,039 0,041 0,005 0,005
GAM Star Japan
Leaders USD 0,001 0,004 0,016 0,002 0,003 -0,001 0,004 Matthews Japan Fund 0,007 0,004 0,010 0,009 0,011 0,007 0,003 Threadneedle Japan Retail 0,011 0,010 0,015 0,021 0,020 0,010 0,009 Schroder ISF Japanese Opportunities 0,007 0,002 0,040 0,043 0,049 0,001 0,001 T Rowe Price Japan Fund 0,006 0,004 0,014 0,018 0,014 0,004 0,003 LF Morant Wright Japan 0,001 0,002 0,004 -0,000 0,001 0,002 0,002 Hennessy Japan Fund -0,001 0,001 0,021 -0,032 -0,025 0,002 0,001 Aviva Investors Japon 0,001 0,001 0,033 0,037 0,029 -0,002 0,001 Handelsbanken Japan Tema 0,005 0,009 0,023 0,014 0,013 0,006 0,009
GAM Star Japan
33
Interfund Equity
Japan 0,009 0,004 0,029 0,046 0,033 0,004 0,004
Standard Life Inv
Japan 0,013 0,011 0,011 0,001 0,001 0,016 0,013 AXA Framlington Japan 0,004 0,004 0,008 0,011 0,008 0,004 0,004 Scottish Widows Japan Growth 0,006 0,005 0,006 0,005 0,004 0,006 0,005 GAM Multistock - Japan 0,005 0,005 0,013 0,004 0,009 0,005 0,005 Lansforsakringar Japanfond -0,001 0,004 0,030 -0,019 0,004 0,003 0,006 Fidelity Funds - Japan 0,007 0,007 0,016 0,007 0,007 0,008 0,007 Fonditalia Equity Japan 0,010 0,005 0,029 0,046 0,033 0,005 0,006 Comgest Growth Japan 0,005 0,006 0,035 0,016 0,012 0,003 0,003 BGF Japan Small & MidCap Opportunities 0,006 0,005 0,010 0,005 0,010 0,007 0,006 AS SICAV I - Japanese Equity 0,002 0,002 0,005 0,001 0,004 0,002 0,002 T Rowe Price Japanese Equity 0,004 0,003 0,024 0,005 0,006 0,003 0,003 Candriam Equities L Japan 0,009 0,005 0,052 0,074 0,105 0,002 0,005
SLI Glo SICAV
Japanese Equities 0,014 0,014 0,056 0,061 0,094 0,008 0,011 Goldman Sachs Japan Equity 0,015 0,010 0,047 0,062 0,079 0,008 0,009 JPMorgan Japan Equity 0,018 0,015 0,053 0,042 0,033 0,013 0,013 BCV Japac 0,004 0,003 0,008 0,006 0,004 0,003 0,002 BNP Paribas Japan Equity -0,001 0,002 0,036 -0,056 -0,032 0,007 0,002 SEB Japanfond 0,003 0,005 0,020 0,005 0,005 0,003 0,005 Fidelity Funds - Japan Advantage 0,006 0,007 0,015 -0,000 -0,001 0,008 0,008 Legg Mason IF Japan Equity 0,001 0,000 0,007 -0,004 0,001 0,002 0,000
Fidelity Inst Japan 0,008 0,005 0,013 - - 0,007 0,006
34
References
Agyei- Ampomah, S. (2007). The post- cost profitability of momentum trading strategies: further evidence from the UK. European Financial Management, 13(4), 776-802.
Avramov, D., Chordia, T., Jostova, G., & Philipov, A. (2007). Momentum and credit rating. The Journal of Finance, 62(5), 2503-2520.
Badrinath, S.G., Gay, S.D., and Kale, J.R., (1989). Patterns in institutional investment, prudence, and the managerial safety-net hypothesis. The Journal of Risk and Insurance, 56, 605-629.
Badrinath, S.G., Wahal, S., (2002). Momentum trading by institutions. Journal of Finance, 57(6), 2449–2478
Baker, H. K., Veit, E. T., & Powell, G. E. (2001). Factors influencing dividend policy decisions of Nasdaq firms. Financial Review, 36(3), 19-38.
Barber, B. Lehavy, R. McNichols, M. and Trueman, B. (2003) Reassessing the returns to analysts' stock recommendations. Financial Analysts Journal, 59 (2) (2003), pp. 88-96.
Barberis, N., Shleifer, A., and Vishny, R.W., (1998). A model of investor sentiment. Journal of Financial Economics, 49, 307-343.
Barroso, P., Santa-Clara P. (2015). Momentum has its moments. Journal of Financial Economics, 116,111- 120.
Bennett, J., R. Sias, and L. Starks, (2003). Greener pastures and the impact of dynamic institutional preferences. Review of Financial Studies, 16, 1203-1239.
Bhojraj, S., & Swaminathan, B. (2006). Macromomentum: returns predictability in international equity indices. The Journal of Business, 79(1), 429-451.
35
Black, F. (1993) ‘Beta and return’, Journal of Portfolio Management 20: 8–18
Bosworth, B., & Flaaen, A. (2009). America’s financial crisis: the end of an era.
Carhart, Mark M., (1997). On Persistence in Mutual Fund Performance. Journal of Finance, 1997, 57–82.
Carhart, M. M., Carpenter, J. N., Lynch, A. W., & Musto, D. K. (2002). Mutual fund survivorship. The review of financial studies, 15(5), 1439-1463.
Chan, L. K., Jegadeesh, N., & Lakonishok, J. (1996). Momentum strategies. The Journal of Finance, 51(5), 1681-1713.
Chen, J., H. Hong, M. Huang and J. Kubik (2004), Does fund size erode performance? Liquidity, organizational diseconomies and active money management. American Economic Review, Volume 94, pp. 1276-1302.
Chui, A. C., Wei, K. C., & Titman, S. (2000). Momentum, legal systems and ownership structure: An analysis of Asian stock markets. Sheridan, Momentum, Legal Systems and Ownership Structure: An Analysis of Asian Stock Markets (December 2000).
Chui, A., Titman, S., and Wei, J., (2010). Individualism and momentum around the world. Journal of Finance, 65, 361–392.
Collins, H., Evans, R., & Gorman, M. (2007). Trading zones and interactional expertise. Studies in History and Philosophy of Science Part A, 38(4), 657-666.
Conrad, J., & Kaul, G. (1998). An anatomy of trading strategies. The Review of Financial Studies, 11(3), 489-519.
36
Daniel, K., Hirshleifer, D., and Subrahmanyamm, A., (1998). Investor psychology and security market under- and overreactions. Journal of Finance, 53, 1839-1885.
Daniel, K. D., & Moskowitz, T. J. (2013). Momentum crashes. Swiss Finance Institute Research Paper, (13-61), 14-6.
De Bondt, W. F., & Thaler, R. (1985). Does the stock market overreact?. The Journal of finance, 40(3), 793-805.
De Bondt, W., Thaler, R., (1987). Further Evidence on Investor Overreaction and Stock Market Seasonality. Journal of Finance, 42 (3), 557-581.
De Haan, L., Kakes, J., (2011) Momentum or contrarian investment strategies: evidence from Dutch institutional investors. Journal of Banking Finance, 35(9), 2245–245.
De Long, B.J., Shleifer, A., Summers, L., and Waldmann ,R., (1990). Positive feedback investment strategies and destabilizing rational speculation. Journal of Finance, 45 (2), 379-395.
Del Guercio, D., (1996). The distorting effect of the prudent-man laws on institutional equity investment. Journal of Financial Economics, 40, 3-62.
Doukas, J. A., & McKnight, P. J. (2005). European momentum strategies, information diffusion, and investor conservatism. European Financial Management, 11(3), 313-338.
Dutt, T., & Humphery-Jenner, M. (2013). Stock return volatility, operating performance and stock returns: International evidence on drivers of the ‘low volatility’anomaly. Journal of Banking & Finance, 37(3), 999-1017.
37
Evans, R. (2007). The incubation bias. Unpublished paper, University of Virginia.
Evans R., (2010). Mutual Fund Incubation. Journal of Finance, 65 (4), 1581-1611.
Falkenstein, E.G., (1996). Preferences for stock characteristics as revealed by mutual fund portfolio holdings. Journal of Finance, 51, 111-135.
Fama, E.F., (1965). The behaviour of stock market prices. Journal of Business, 38, 34-105.
Fama, E. F., & French, K. R. (1996). Multifactor explanations of asset pricing anomalies. The journal of finance, 51(1), 55-84.
Fama, Eugene F., (1998), Market efficiency, long-term returns and behavioral finance, Journal of Financial Economics 49, 283–306
Fama, E. F., & French, K. R. (2008). Dissecting anomalies. The Journal of Finance, 63(4), 1653-1678.
Fama, E., French, K., (2012). Value, size, and momentum in international stock returns. Journal of Financial Economics, 105, 457–472.
Ferreira, M. A., Keswani, A., Miguel, A. F., & Ramos, S. B. (2012). The flow-performance relationship around the world. Journal of Banking & Finance, 36(6), 1759-1780.
Fiore, C., & Saha, A. (2015). A Tale of Two Anomalies: Higher Returns of Low- Risk Stocks and Return Seasonality. Financial Review, 50(2), 257-273.
Fuertes, A. M., Miffre, J., & Fernandez- Perez, A. (2015). Commodity strategies based on momentum, term structure, and idiosyncratic volatility. Journal of Futures Markets, 35(3), 274-297.
38
Gompers, P., Metrick, A., (2001). Institutional investors and equity prices. Quarterly Journal of Economics, 116, 229-260.
Grundy, B. D., & Martin, J. S. M. (2001). Understanding the nature of the risks and the source of the rewards to momentum investing. The Review of Financial Studies, 14(1), 29-78.
Grinblatt, M., & Titman, S. (1992). The persistence of mutual fund performance. The Journal of Finance, 47(5), 1977-1984.
Grinblatt, M., Titman, S., (1993). Performance Measurement without Benchmarks: An Examination of Mutual Fund Returns. Journal of Business, 1993, 66(1), 47-68.
Grinblatt, M., Titman, S., Wermers, R., (1995). Momentum Investment Strategies, Portfolio Performance, and Herding: A Study of Mutual Fund Behavior. American Economic Review, 85 (5), 1088- 1105.
Grinblatt, M., Jostova, G., Petrasek, L., & Philipov, A., (2016). Style and skill: hedge funds, mutual funds, and momentum. Mutual Funds, and Momentum (July 16, 2016).
Griffin, J.M., Ji, X., Martin, J.S., (2003). Momentum investing and business cycle risk: evidence from pole to pole. The Journal of Finance, 58(6), 2515–2547.
Gupta, K., Locke, S., & Scrimgeour, F. (2013). Profitability of momentum returns under alternative approaches. International Journal of Managerial Finance, 9(3), 219-246.
Hameed, A., & Kusnadi, Y. (2002). Momentum strategies: Evidence from Pacific Basin stock markets. Journal of financial research, 25(3), 383-397.
Hirshleifer, D. (2001). Investor psychology and asset pricing. The Journal of Finance, 56(4), 1533-1597.
39
Hong, H., Stein, J.C., (1999). A unified theory of underreaction, momentum trading and overreaction in asset markets. Journal of Finance, 54, 2143-2184.
Jegadeesh, N., Titman, S., (1993). Returns to Buying Winners and Selling Losers: Implications for Stock Market Efficiency. The Journal of Finance, 48 (1), 65-91.
Lakonishok, J., Shleifer, A., and Vishny, R.W., (1992). The impact of institutional trading on stock prices. Journal of Financial Economics, 32, 23-43.
Lee, C. M., & Swaminathan, B. (2000). Price momentum and trading volume. the Journal of Finance, 55(5), 2017-2069.
Lesmond, D. A., Schill, M. J., & Zhou, C. (2004). The illusory nature of momentum profits. Journal of financial economics, 71(2), 349-380.
Li, J., Yao, Y., Chen, Y., & Lee, C. F. (2018). Option prices and stock market momentum: evidence from China. Quantitative Finance, 18(9), 1517-1529.
Lo, A. W., H. Mamaysky, and J. Wang. (2000). “Foundations of Technical Analysis:
Computational Algorithms, Statistical Inference, and Empirical Implementation.” The Journal of Finance 55 (4): 1705–1765.
Lui, W., Strong, N., & Xu, X. (1999). The profitability of momentum investing. Journal of Business Finance & Accounting, 26(9- 10), 1043-1091.
Kambouroudis, D. S., & McMillan, D. G. (2015). Is there an ideal in-sample length for
forecasting volatility?. Journal of International Financial Markets, Institutions and Money, 37, 114-137.
40
Kaminsky, G., Lyons, R., Schmukler, S., (2011). ‘Mutual fund investment in emerging markets: an overview’, World Bank Economic Review, 15 (2), 315-340
Karoui, A., Meier, I., (2009). Performance and characteristics of mutual fund starts. The European Journal of Finance, 15, 487-509.
Korajczyk, R. A., & Sadka, R. (2004). Are momentum profits robust to trading costs?. The Journal of Finance, 59(3), 1039-1082.
MacKinlay, C. A. (1995) ‘Multifactor models do not explain deviations from the CAPM’, Journal of Financial Economics 38: 3–28.
Malkiel, B. G., & Fama, E. F. (1970). Efficient capital markets: A review of theory and empirical work. The journal of Finance, 25(2), 383-417.
Moskowitz, T., Grinblatt, M. (1999). Do industries explain momentum. Journal of Finance, 54(4): 1249-1290
Moskowitz T., Ooi Y., Pedersen L., (2012). Time series momentum. Journal of Financial Economics, 104, 228–250.
Mulvey, J., & Kim, W. (2008). Constantly rebalanced portfolios-is mean-reverting necessary. Technical report, working paper Princeton University, Princeton, NJ, USA.
Nofsinger, J., Sias. R., (1999). Herding and Feedback Trading by Institutional and Individual Investors.” Journal of Finance, December 1999, 2263–2295.
Novy-Marx R., (2012). Is momentum really momentum?. Journal of Financial Economics, 103, 429 - 453.
41
Sias, R., Starks, L., Titman, S., (2001). The Price Impact of Institutional Trading. Unpublished working paper.
Tse, Y. (2015). Momentum strategies with stock index exchange-traded funds. The North American Journal of Economics and Finance, 33, 134-148.
Titman, S., & Grinblatt, M. (1989). Mutual fund performance: An analysis of quarterly portfolio holdings. Journal of Bussines, 62(3).
Titman, S. (2001). The Modigliani and Miller theorem and market efficiency (No. w8641). National Bureau of Economic Research.
Wermers, R. (1999) Mutual fund herding and the impact on stock prices. Journal of Finance 54:581–622.
Wermers, R., Yao, T., & Zhao, J. (2007). The investment value of mutual fund portfolio disclosure. Unpublished working paper, University of Maryland, 8, 47-59.
