Download Momentum and Investor Sentiment

Survey
yes no Was this document useful for you?
   Thank you for your participation!

* Your assessment is very important for improving the workof artificial intelligence, which forms the content of this project

Document related concepts

Stock wikipedia , lookup

Algorithmic trading wikipedia , lookup

Short (finance) wikipedia , lookup

Securities fraud wikipedia , lookup

Technical analysis wikipedia , lookup

Index fund wikipedia , lookup

Stock market wikipedia , lookup

Stock exchange wikipedia , lookup

2010 Flash Crash wikipedia , lookup

Efficient-market hypothesis wikipedia , lookup

Stock selection criterion wikipedia , lookup

Market sentiment wikipedia , lookup

Transcript
Momentum and Investor Sentiment : Evidence from Asian Stock Markets
Shangkari V.Anusakumar
Universiti Sains Malaysia
School of Management
USM, 11800 Penang, MALAYSIA
[email protected]
Tel: +(604) 6533984
Ruhani Ali
Universiti Sains Malaysia
Graduate School of Business
USM, 11800 Penang, MALAYSIA
[email protected]
Tel: +(604) 6533954
Hooy Chee Wooi
Universiti Sains Malaysia
School of Management
USM, 11800 Penang, MALAYSIA
[email protected]
Tel: +(604) 6532897
Momentum and Investor Sentiment : Evidence from Asian Stock Markets
Abstract
The study provides the first evidence on the effect of investor sentiment on momentum
profitability in Asia. A sample of 13 Asian countries is analysed over a period of 12 years
from January 2000 to December 2011. We find that momentum arises only during optimistic
and mild periods. Momentum is conspicuously absent for periods of pessimism. The evidence
suggests that investors are detail oriented during pessimistic periods and thereby hinder the
occurrence of momentum in the stock market. In addition to local sentiment, we also find that
global sentiment affects momentum which affirms the contagious nature of sentiment. The
results are robust to changes in sentiment period classification and use of alternative proxies.
Moreover, we also find that the findings are unaffected after taking into consideration firm
size and trading volume.
Keywords: Momentum, Sentiment, Global Sentiment, Asian markets
1
1
Introduction
Within the large body of literature that documents return predictability, the momentum effect
is arguably one of the most important and intriguing. It is one the few anomalies that have yet
to be explained in its entirety. In fact, Fama (1998) acknowledged momentum as one of the
most difficult anomalies to account for. The momentum effect was first documented by
Jegadeesh and Titman (1993). The authors demonstrated that stocks which performed poorly
(well) in the past continue to perform poorly (well) in the future. The basic concept of
momentum strategy is to buy ‘winners’ (stocks that performed well in the past) and sell
‘losers’ (stocks that performed poorly in the past).
Sentiment is proposed as one of the factors that could affect the levels of momentum.
Investor sentiment, as proxied by consumer confidence index, has been shown to influence
future stock returns (Schmeling, 2009). Higher sentiment is linked to a lower stock returns in
the future. However, the predictive power of sentiment ranges from strong to none depending
on the stock market being investigated. A variety of studies have surged linking sentiment
and other financial aspects from IPO prices to feedback trading (e.g. Ben-Rephael, Kandel &
Wohl, 2011; Chau, Deesomsak & Lau, 2011; Cornelli, Goldreich & Ljungqvist, 2006;
Küçükaslan & Çelik, 2010; Liao, Huang & Wu, 2011). For momentum, Antoniou, Doukas
and Subrahmanyam (2012) found higher momentum during periods of high investor
sentiment compared to low sentiment and that sentiment is absent during pessimistic periods.
As highlighted by Schmeling (2009), the effect of sentiment varies from country to country,
and as such the relationship between momentum and sentiment needs to be reexamined in
Asian markets.
Investor sentiment was found to be related to momentum in the US (Antoniou et al.,
2012) but this relationship may or may not hold in Asia. Firstly, manifestation of momentum
in Asia drastically differs from other regions around the world (Griffin et al., 2003). So much
2
so that Hameed and Kusnadi (2002) argued that the factors that drive momentum in Asia may
not be the same as those in the US. More importantly, the psychology of Asians are notably
distinct from Westerners including reasoning (Buchtel & Norenzayan, 2008) and modes of
thinking (Nisbett, 2003). Hedden, Ketay, Aron, Markus, and Gabrieli (2008) showed that
individuals from America and Asia have distinct brain activity patterns when exposed to the
same visual problems. Functional magnetic resonance imaging (fMRI) scans revealed higher
brain activity for Americans when solving problems involving relative judgement compared
to absolute judgment. On the other hand, the reverse was true for East Asians. Given these
clear distinctions, the question arises as to whether the sentiment would still be related to
momentum in Asia and if so what would be the nature of the relationship.
To the best of our understanding, the first and thus far the only study specifically on
sentiment and momentum was conducted by Antoniou et al. (2012) for the US market.
Needless to say, the literature is very sparse. International evidence in support or against the
existing results would provide much needed insight into the issue. In addition to this, global
sentiment and sentiment during portfolio holding period is also hypothesized to have an
effect on momentum returns and is investigated accordingly. Not only does this thesis
provide out-of-sample evidence, it also expands the study of sentiment to incorporate global
investor sentiment.
Sentiment represents the state of mind of the investors. From a psychological
viewpoint, investors’ decision to buy, sell or hold would be influenced by the investor’s
psychological state at that point in time. Decision making is not undertaken in isolation, thus
emotional state could sway trading behaviour of investors. The effect would then translate
into either a detrimental or additive factor on the propensity to commit cognitive biases. In a
positive state, individuals are more likely to stick to their normal routine but negative state
elicits a more severe response whereby processing is more detail oriented (Schwarz, 2002).
3
Ali and Gurun (2009) also echoed the view that optimism decreases the attentiveness of
investors. To surmise, individuals are more alert during pessimistic periods and less attentive
at optimistic times. Inattentiveness causes a delayed reaction to the arrival of new information
supporting behavioural theories of underreaction (Dellavigna & Pollet, 2009). As for
pessimistic periods, the increased awareness and processing of information could reduce or
even remove the cognitive bias that drives momentum. Thus, it could be conjectured that
optimism leads to a higher level of momentum whereas pessimism is associated with lower or
even absence of momentum.
In addition to sentiment in the local stock market, cognitive biases and
correspondingly momentum returns could be swayed by sentiment on an international scale.
Baker, Wurgler and Yuan (2012) also advocated the view and demonstrated the influence of
global sentiments on stock prices. Moreover the authors suggested that 'sentiment is
contagious across markets'. Chang, Faff and Hwang (2009) suggested that US investor
sentiment may reflect global sentiments. Investigating 38 countries, the authors documented
strong sentiment contagion. Thus it is not a stretch to postulate that global investor sentiment,
apart from local sentiment, would affect stock momentum in Asia.
The remainder of the paper is organised as follows. We describe the data source and
methodology in Section 3 and 4 respectively. In Section 5, we provide the results and
discussion. Finally Section 5 provides the conclusion.
2
Data
The study encompasses a 12-year period from 1 January 2000 to 31 December 2011. This
particular period is examined in the interest of using up to date information to reflect current
market conditions. The premise of this study is Asian stock markets. As the sample time
period covers the years 2000 to 2011, stock exchanges newly founded for this duration could
not be included in the sample. This excludes stock exchanges established from 2000 onwards
4
such as the Hochiminh Stock Exchange (HOSE) in Vietnam. As in Griffin et al. (2005), a
minimum of 50 stocks are required to be listed on the stock exchange to enable momentum
and subsequent tests to be carried out. This restriction weeds out exchanges with limited
listings such as the Maldives Stock Exchange (MSE). The screening results in a final sample
of stock exchanges from 13 Asian countries: Bangladesh, China, Hong Kong, India,
Indonesia, Japan, Malaysia, Pakistan, Philippines, Singapore, South Korea, Taiwan and
Thailand. Stock return index is obtained from Datastream. In addition to stock return data,
trading volume, firm size, consumer confidence indices and other related information are also
obtained from Datastream. Table 1 states the sample countries used in this study and also
presents the total number of sample stocks for each country. This figure includes active and
‘dead’ (delisted) stocks. The number of stocks available at each juncture for portfolio
formation and holding differs from this figure.
The consumer confidence index has been frequently used a proxy for investor
sentiment (Lemmon & Portniaguina, 2006). The sentiment index is constructed based on
survey information. The questionnaire and frequency of data collection differs from country
to country. Nevertheless it is a reliable and independent source for the measurement of
sentiments. An advantage of using the consumer confidence index is the fact that the values
are derived independently of the stock market. Measures derived from stock market related
data could be compounded by a multitude of factors. Therefore a proxy independent of the
stock market is needed. Consumer confidence index provides such a measure as it is based on
direct survey of individual consumers and is thus free from stock market related factors.
Moreover as stated by Schmeling (2009), consumer confidence index “seems to be the only
consistent way to obtain a sentiment proxy that is largely comparable across countries” (p.
397). Consumer confidence index available for each country is obtained from Datastream.
5
Following Chang et al. (2009), global sentiment is proxied by US investor sentiment.
For this purpose, a US based consumer confidence index is used to gauge the global investor
sentiments. The Conference Board Consumer Confidence Index has been used to measure
investor sentiment in several recent US market based studies (e.g. Ho and Hung, 2009; Tang
and Yan, 2010). Moreover, Qiu and Welch (2004) and Lemmon and Portniaguina (2006)
have noted that the consumer confidence index is appropriate measure of investor sentiment.
The index is formed based on a monthly survey of 5000 households in the US.
3
Methodology
The 6 month formation and 6 month holding period strategy has been found to be
consistently profitable (Jegadeesh & Titman, 1993) and is frequently used in literature.
Stocks are ranked based on cumulative returns from t-2 to t-7. Equally weighted winner and
loser portfolios are formed using the top (winner stocks) and bottom (loser stocks) 10% of the
stocks. A month is skipped after portfolio formation in order to mitigate microstructure biases.
The constituents of the winner and loser portfolio are maintained for 6 months. The monthly
returns for the winner, loser and momentum portfolio are computed for each month from t to
t+5. The procedure is repeated for each month.
At the end of formation period t, the weighted rolling average consumer confidence
index of the previous 3 months is calculated with the weight of 3, 2 and 1 for month t, t-1 and
t-2 respectively.
AvgSent =
1
2
3
Sent t − 2 + Sent t −1 + Sent t
6
6
6
AvgSent is the weighted average sentiment used to classify formation periods as pessimistic
or optimistic. Sentt-2, Sentt-1 and Sentt represent confidence index value at month t-2, t-1 and t
respectively.
6
A particular formation period's sentiments is high/optimistic (low/optimistic) when it
ranks in the top (bottom) 30% of the average sentiment values. The remaining portfolios are
assumed to have been formed during a ‘mild’ period i.e. neither high nor low level of
sentiment. The cut off is the same as that employed by Antoniou et al. (2012). Periods of high
and low investor sentiment are thus defined on a relative basis for the entire sample period.
Following this, the momentum returns for the portfolios formed during pessimistic, mild and
optimistic periods are assessed. For local sentiment, the consumer confidence index for each
country is used for the aforementioned analysis. Apart from local sentiment, the stock market
returns could be swayed by global sentiments. Stock markets are no longer closed entities;
the participants of stocks markets are comprised of a broad range of investors from various
countries. Global sentiment has been shown to effect stock prices apart from local sentiment
(refer to Baker et al., 2012; Chang et al., 2009). For global sentiment, the procedure is
performed using Conference Board Consumer Confidence Index.
4
4.1
Empirical Results
Momentum Returns for Asian Markets
Table 2 presents the average monthly returns along with corresponding t-statistics for winner,
loser and momentum portfolio for the 13 countries. As in Chui et al. (2010), the final row
reports the averages. The portfolios are formed monthly by sorting the stocks based on the
past 6 month returns. The stocks in the best performing decile is designated as winners and
worst decile are designated as losers. Winner and loser portfolios are then formed using the
selected stocks. Momentum portfolio is constructed by taking a long position in the winner
stocks and a short position in the loser stocks. Winner, loser and momentum portfolios are
held for the following six months after skipping one month to eliminate market
7
microstructure biases (as is the convention in momentum studies). Momentum returns are
computed as the differential between winner and loser portfolio returns.
The winner portfolio generates positive return for all of the countries. The returns are
statistically significant for a majority of the countries, specifically for nine out of the thirteen
the countries. This provides evidence of significant return continuations for winner stocks. In
other words, stocks that performed well in the past continue to perform well in the future. In
contrast, loser portfolio returns are significant for only six countries. Returns to the
momentum portfolio are generally positive. Out of the sample of 13 countries, 11 countries
have positive returns for the momentum portfolio while 2 countries have negative returns.
The highest momentum can be observed for Bangladesh whilst Philippines has the lowest
return. Roughly one third of the sample countries display statistically significant momentum.
Certain countries exhibit a high degree of momentum comparable to those reported in the US
market. In short, there is evidence of momentum profitability in selected Asian countries.
Bangladesh, in particular, has markedly strong momentum in the stock market. The
momentum portfolio earns 1.470% per month which is higher than the returns reported in the
US (e.g. Jegadeesh and Titman, 2001). The winner and loser portfolios both generate
significant returns on their own. In particular, the winner portfolio has noticeably strong
returns. The findings of momentum in Bangladesh concur with the results of Chui et al.
(2011). As indicated by the last row of Table 2, on average, there is significant momentum in
Asia at the 5 percent level.
4.2
Momentum and Local Sentiment
Table 3 reports the portfolio returns for the momentum strategy during three sentiment states:
optimistic, mild and pessimistic. Sentiment states classification is done based on the weighted
average of local consumer confidence index values over the portfolio formation period. The
8
average monthly returns, in percentages, are presented for the winner, loser and momentum
portfolio along with the associated t-statistics. Due to unavailability of local sentiment data,
the effect of local sentiment on momentum could not be explored for Bangladesh, Singapore
and Pakistan.
As can be observed from Table 3, the non-existence of momentum for pessimistic
periods is glaringly obvious. As hypothesized by Antoniou (2012), none of the thirteen
countries display any momentum for the pessimistic period. Momentum exists exclusively in
the optimistic and mild periods. Further solidifying the failure of momentum strategy during
pessimistic times, there are instances where returns for the pessimistic period are outright
negative whereas optimistic periods have strong positive returns. For example, the strategy in
the Japanese markets undergoes statistically significant losses of 2.599% on a monthly basis
for pessimistic period. At the other end of the spectrum, the momentum portfolio in the
optimistic period garners a substantial monthly return of 1.280% which is statistically
significant at the 1 percent level. This highlights the stark variation in momentum between
the sentiment states.
Another notable finding is the presence of momentum during optimistic and mild
periods in countries where otherwise momentum could not be found. Indonesia, Japan and
Taiwan have no momentum for the overall strategy but record high and statistically
significant returns to the momentum portfolio for the optimistic and/or mild period. Thus, it
could be conjectured that sentiment state is one the factors causing the apparent lack of
momentum or rather masking the presence of momentum in Asia.
A salient fact emerges from this study; sentiment does influence momentum
profitability. The most striking result can be observed for Japan. Whilst studies have reported
contradictory account of momentum levels in other markets such as Hong Kong, the Japanese
stock market has always been reported to be devoid of momentum and returns are often
9
negative. The earlier part of this thesis also supports the notion that there is no momentum in
Japan. Surprisingly, there are high levels of momentum once sentiment in taken into account.
Returns for momentum strategy is high during optimistic and mild periods but is downright
unprofitable during pessimistic periods. There is a staggering return differential of 3.879%
between optimistic and pessimistic periods. The evidence suggests that sentiment could be
the reason for the absence of momentum in Japan. A similar account could be noted for
Taiwan, another country where momentum has been chronically non-existent.
Overall, the evidence on local sentiment concurs with the findings of Antoniou et al.
(2012); momentum is only present for high sentiment periods. Momentum portfolio returns
are in part derived from the poor performance of the loser portfolio. The factor that
differentiates high and low sentiment periods is largely the loser portfolio. Loser portfolio
returns are higher during pessimistic periods compared to optimistic periods. Moreover,
returns for loser portfolio are on par with or higher than winner portfolio for low sentiment
states. It is this fact that causes the absence of momentum during pessimistic periods. The
evidence suggest that investor’s have a greater propensity to engage in detailed processing
during low sentiment periods which causes the elimination of momentum during pessimistic
periods (Schwarz, 2002).
4.3
Momentum and Global Sentiment
Table 4 reports the winner, loser and momentum portfolio returns, along with the t-statistics,
during periods of varying global sentiment. The sentiment periods are split into optimistic,
mild and pessimistic based on the past index values that rank in the top 30%, middle 40% and
bottom 30% respectively. The portfolio return figures are in percentage and represent the
average monthly return.
10
Momentum portfolio returns are positive for all countries during the optimistic period,
out of which five countries have significant returns. For the mild period, six countries have
significant returns whilst there are no significant returns for the pessimistic period. Overall,
eight of the thirteen countries have significantly positive returns to the momentum portfolio
during the optimistic and/or mild period. The highest momentum portfolio return reaches up
to 2.904% per month. In other words, more than half of the Asian countries exhibit strong
momentum.
One of the main and apparent finding is that the pessimistic period is devoid of
momentum. In this aspect, the evidence reconciles with the findings of the preceding section
where local investor sentiment was investigated. Furthermore, almost all of the countries,
eleven of the thirteen countries, have negative returns to the momentum portfolio. This
presents overwhelming evidence on the failure of momentum strategy during periods of low
sentiment. In contrast, momentum strategy fares better for the other sentiment states
especially optimistic periods. Momentum is undoubtedly the highest during optimistic
periods. Moreover countries such as Bangladesh and Pakistan have significant returns only
during optimistic periods. Momentum in optimistic periods outperforms pessimistic periods
primarily due to the loser portfolio. For the optimistic periods, the loser portfolio performs
poorly as expected. However the loser portfolio performs comparatively better in during
pessimistic periods. For example, Bangladesh has a measly loser portfolio return of 0.035%
for optimistic period but the fortunes of the loser portfolio changes drastically during
pessimistic period as the average monthly return is 2.880% and is marginally significant.
Last but not least, there are countries which display strong momentum for high
sentiment states but do not have any momentum for the overall momentum strategy. This
evidence is again similar to that found for the local investor sentiment. China, Singapore and
Thailand did not have any significant momentum for the unrestricted strategy (Table 2).
11
Upon segregating momentum based on global sentiment, these countries display high levels
of momentum. For instance, China has statistically significant monthly return of 1.288%
(1.019%) for the optimistic (mild) period. However for the overall strategy China had an
insignificant and comparatively meager return of 0.555% per month (Table 2).
Global sentiment appears to have a stronger effect on momentum than local sentiment.
An extreme example of this is the Chinese stock market. Momentum in China is unaffected
by local sentiment i.e. there is no momentum in the market regardless of the sentiment state.
However significant returns emerge when global sentiment is used for the analysis. As the
evidence suggests that global sentiment may have an equal if not greater influence on
momentum profitability, investors should pay heed to both local and global sentiment when
implementing in the momentum strategy in Asian markets and in some cases more to the
latter.
In a nutshell, global sentiment also affects momentum profitability. This finding
corroborates with the contagious nature of sentiment as noted by Baker et al. (2012). It is has
been suggested that sentiment spreads rapidly through mass media (Du, 2010). Baker et al.
(2012) suggested that “capital flows are a key mechanism through which global sentiment
develops and propagates, but there are surely others, including word-of-mouth and the
media” (pg. 104). Regardless of the means by which sentiment spreads, the fact that global
sentiment affects the level of momentum profitability in Asian markets further confirms the
contagious nature of investor sentiment.
4.4
Alternative Sentiment Classification
The investor sentiment investigations thus far have been conducted by classifying sentiment
period based on a 30% cut off. A period is optimistic (pessimistic) if the index value is in the
top 30% (bottom) of the time series of sentiment index values. Remaining periods are
designated as mild periods. In this section, an alternative sentiment cut-off is selected and
12
investigated to ensure the findings of this study are robust to changes in sentiment
classification. Specifically the analysis is repeated using a 40% cut-off for sentiment
classification. Table 5 reports the results of this analysis1.
For local sentiment, the findings are similar with the use of a 40% cutoff for sentiment
classification instead of 30% cut off. Pessimistic periods are devoid of momentum; none of
the thirteen countries exhibit statistically significant returns to the momentum portfolio
during pessimistic periods. During optimistic and mild periods, there are countries that
display high levels of profitability. For example, Japan has a return of 1.025% during
optimistic periods, which is comparable to the momentum levels reported in US. In contrast,
momentum portfolio return in Japan during pessimistic period is an insignificant -1.239%. In
short, there’s a return difference of 2.264% between optimistic and pessimistic period.
Investigation using 40% cut off for global sentiment also echoes the earlier findings.
There is a clear divide between profitability during pessimistic periods and periods with
higher sentiment. Returns to the momentum strategy are rampantly negative for pessimistic
periods. Moreover the negative return for Japan is statistically significant indicating that an
investor implementing a momentum strategy during global pessimistic periods would stand to
a 1.176% per month. In contrast, an investor implementing the strategy during optimistic
periods could gain an average monthly return of 1.341%. High momentum profitability could
be found in selected countries during high sentiment periods. Bangladesh, for example, offers
a high momentum portfolio return of 2.101% (significant at 5 percent level) for optimistic
periods. Overall, the earlier findings are generally intact irrespective of the sentiment
classification.
1
For brevity, only optimistic and pessimistic momentum returns are reported. Full results are available upon
request.
13
4.5
Alternative Sentiment Proxies
For robustness, we repeat the sentiment analysis using an alternative sentiment measure,
University of Michigan Sentiment Index. In addition to the survey based sentiment measure,
sentiment index by Baker and Wurgler (2006) which is a market based measure is also tested
in this section. Table 6 reports the results of the study using the alternative sentiment
measures.
Firstly, results for the analysis using the University of Michigan Sentiment Index is
presented. Eight of the thirteen countries have significant momentum during optimistic and/or
mild periods. Momentum strategy performs poorly during pessimistic periods as there are no
significant returns; the returns are also largely negative. The failure of momentum strategy
during pessimistic periods is clearly visible. Thus, the earlier conclusion that sentiment
affects momentum remains unaltered with the use of the alternative sentiment proxy,
University of Michigan Sentiment Index.
For the Baker and Wurgler (2006) index, the results are sporadic. There are only two
countries with momentum during optimistic periods. Moreover, only marginally significant
momentum is present for mild period whereas there are highly significant momentum for two
countries and one marginally significant negative return during pessimistic period. There
appears to be no detectable pattern. Given that momentum is equally present in optimistic
periods and pessimistic periods, this leads to the conclusion that momentum is not affected by
sentiment as measured by the Baker and Wurgler (2006) index. The lack of evidence could be
due to the nature of the Baker and Wurgler (2006) index. The index is derived from variables
from the stock market such as returns on first day returns on IPO. On the other hand,
Conference Board Consumer Confidence Index and University of Michigan Consumer
Sentiment Index are survey based measure where the index is constructed based on direct
14
survey of consumers. The results suggest that the Baker and Wurgler (2006) index may not
be able to fully capture the components of global sentiment.
An alternative survey based measure, University of Michigan sentiment index,
produces a similar, if not stronger results. On the other hand, the analysis with Baker and
Wurgler (2006) composite index yields sporadic momentum across the sentiment states with
no obvious pattern. This is at odds with the results of Antoniou et al. (2012) where results
were similar whether consumer confidence index by Conference Board (proxy used
throughout the study) or the Baker and Wurgler (2006) composite index (alternative proxy
tested for robustness) were used for the analysis. It is possible that the while Baker and
Wurgler (2006) composite index is an sufficient proxy for the local US market, the index
poorly captures global sentiment. Moreover, Conference Board consumer confidence index
and University of Michigan sentiment index is a survey based proxy of sentiment which may
provide an edge over the composite index which is derived from market based variables.
4.6
Momentum, Sentiment and Size
In this section, the robustness of the effect of sentiment on momentum is further tested by
analyzing size. The effect of local sentiment on momentum for the size categories is reported
in Table 7. The effect of investor sentiment on momentum persists irrespective of the firm
size. In each size category, large, medium and small, a distinct pattern can be observed across
the sentiment states; momentum is strong in optimistic and mild period whereas pessimistic
periods have little or no momentum. In fact, the returns to the momentum portfolio are
generally negative in the pessimistic period for the three size categories. This is readily
apparent for Japan where momentum is present only for optimistic and mild periods and turns
significantly negative for pessimistic period. For example, the medium stock category for
Japan has statistically significant returns of 1.020%, 0.772% and -2.531% per month for
optimistic, mild and pessimistic period respectively.
15
The effect of global sentiment on momentum for the size categories is reported in
Table 8. Again, momentum returns for the large, medium and small stocks are all greatly
affected during times of global pessimism. There are primarily negative returns for the
momentum portfolio throughout the three size categories for the pessimistic period. Small
stocks suffer the most as evidenced by the statistically significant negative returns (in two
instances). For example, investors focusing momentum strategy on small stocks stand to lose
a staggering 2.788% per month during globally pessimistic periods. Small stocks do not
perform as well as large and medium stocks generating comparatively lower momentum
portfolio returns. Moreover, the presence of statistically significant momentum is largely
concentrated in large and medium stocks during optimistic and mild periods.
4.7
Momentum, Sentiment and Volume
In this section, the robustness of the effect of sentiment on momentum is further
tested by analyzing trading volume. Trading volume may contain an element of investor
sentiment (Baker & Wugler, 2006). Optimistic investors are more likely to engage in trading
activity in a market with short-sales constraints and this activity is reflected in trading volume
and generally in liquidity (Baker & Stein, 2004). In a sense, trading volume may reflect
investor sentiment but trading volume in itself is a simple and imperfect proxy of sentiment
as it is confounded by factors unrelated to sentiment.
At the end of each formation period, the stocks are segregated into three volume
portfolios; high, medium and low. Then winner, loser and momentum portfolios are formed,
by ranking stocks on the past six month return, within the three volume categories. Finally
the portfolios are classified as optimistic, mild or pessimistic according to the average
confidence index value. The monthly momentum returns for these triple sorted portfolios are
then computed.
16
Table 9 reports the results of this robustness analysis for local sentiment. The results
concur with those of Section 4.2, pessimistic period is largely devoid of momentum while
optimistic and mild period have instances of strong momentum. Momentum in all three
volume categories appear to be affected by local sentiment. For example, high volume stocks
in Japan have a high significant return of 1.733% during optimistic period but this figure
dramatically drops to -2.112 during pessimistic periods. Low volume stocks show a similar
decrease in momentum as sentiment drops with statistically significant returns of 0.855%,
0.765% and -2.049% for optimistic, mild and pessimistic periods respectively. The earlier
remarkable finding of momentum in Japan for high sentiment periods still holds and is
perhaps stronger after taking into account trading volume.
In order to examine the robustness of the effect of global sentiment on momentum
across volume categories, the aforementioned analysis is repeated using the proxy for global
investor sentiment index. Table 10 reports the results of this analysis. Momentum portfolio
returns are all insignificant for global pessimistic periods for all of the volume categories.
The only exception is the medium volume category for China and even then the momentum
portfolio return is negative with marginally significant return of -1.079%, indicating that
investors could face substantial loses during pessimistic periods. On the other hand, there are
momentum portfolio returns as high as 3.546% for global optimistic and mild periods.
Significant momentum can be observed for all three categories. For example, high volume
stocks in South Korea yield 1.495% per month during mild periods whereas returns are
approximately 2% for medium volume stocks; low volume stocks are also profitable with
momentum portfolio returns of 1.610% The evidence indicates that momentum is affected by
global sentiment irrespective of trading volume.
17
5
Conclusion
The central finding of this study is that sentiment affects momentum profitability in Asia.
Momentum is present only during optimistic and mild periods. Pessimistic periods are
fraught with negative returns. More importantly, countries where there is persistent absence
of momentum display significant momentum once sentiment in taken into account. Japan, for
example, has significant momentum during states of high sentiment. On the other hand,
significant negative returns to the momentum portfolio is present during pessimistic periods.
This is what deprives these markets of momentum. In addition to the local sentiment
prevalent in the market, sentiment on a global scale influences momentum. In some cases,
global sentiment appears to have a greater effect on momentum compared to local
momentum. Moreover the findings are robust to changes in the classification of sentiment
periods and alternative sentiment and even after taking into account firm size and trading
volume. Bottom-line is that momentum is absent during pessimistic periods.
The findings provides an interesting revelation to investors. Whilst higher sentiment
periods provide investors with significant momentum portfolio returns, pessimistic periods do
not yield any significant returns and in some cases could even lead to substantial losses.
Investors seeking to implement momentum strategy in Asia and possibly elsewhere should be
cautious of the sentiment prevalent at the time of portfolio formation. Moreover global
sentiment should be taken into consideration as well. Implementing momentum strategy
during consistently pessimistic periods could prove to be disastrous. The message is clear:
investors should stand clear of pessimistic periods. However, it should be noted that trading
costs have not been taken into account and this area could be an interesting consideration for
future papers.
18
References
Ali, A., & Gurun, U. G. (2009). Investor Sentiment, Accruals Anomaly, and Accruals
Management. Journal of Accounting, Auditing and Finance, 24(3), 415 - 431.
Antoniou, C., Doukas, J. A., & Subrahmanyam, A. (2012). Sentiment and Momentum.
Journal of Financial and Quantitative Analysis, Forthcoming.
Baker, M., & Stein, J. C. (2004). Market liquidity as a sentiment indicator. Journal of
Financial Markets, 7(3), 271-299.
Baker, M., & Wurgler, J. (2006). Investor Sentiment and the Cross-Section of Stock Returns.
The Journal of Finance, 61(4), 1645-1680.
Baker, M. P., Wurgler, J. A., & Yuan, Y. (2012). Global, Local, and Contagious Investor
Sentiment. Journal of Financial Economics, 104, 272-287.
Ben-Rephael, A., Kandel, S., & Wohl, A. (2011). Measuring investor sentiment with mutual
fund flows. Journal of Financial Economics, In Press, Corrected Proof.
Buchtel, E. E., & Norenzayan, A. (2008). Which should you use, intuition or logic? Cultural
differences in injunctive norms about reasoning. Asian Journal of Social Psychology,
11(4), 264-273.
Chang, C., Faff, R. W., & Hwang, C.-Y. (2009). Sentiment Contagion, Corporate
Governance, Information and Legal Environments (Working paper).
Chau, F., Deesomsak, R., & Lau, M. C. K. (2011). Investor sentiment and feedback trading:
Evidence from the exchange-traded fund markets. International Review of Financial
Analysis, In Press, Accepted Manuscript.
Chui, A. C. W., Titman, S., & Wei, K. C. J. (2010). Individualism and Momentum around the
World. The Journal of Finance, 65(1), 361-392.
Cornelli, F., Goldreich, D., & Ljungqvist, A. (2006). Investor Sentiment and Pre-IPO
Markets. The Journal of Finance, 61(3), 1187-1216.
Dellavigna, S., & Pollet, J. M. (2009). Investor Inattention and Friday Earnings
Announcements. The Journal of Finance, 64(2), 709-749.
Fama, E. F. (1998). Market efficiency, long-term returns, and behavioral finance. Journal of
Financial Economics, 49(3), 283-306.
Griffin, J., Ji, S., & Martin, S. (2005). Global Momentum Strategies: A Portfolio. Perspective.
Journal of Portfolio Management, 31, 23 - 39.
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.
19
Hameed, A., & Kusnadi, Y. (2002). Momentum Strategies: Evidence from Pacific Basin
Stock Markets. Journal of Financial Research, 25(3), 383-397.
Hedden, T., Ketay, S., Aron, A., Markus, H. R., & Gabrieli, J. D. E. (2008). Cultural
Influences on Neural Substrates of Attentional Control. Psychological Science, 19(1),
12-17.
Jegadeesh, N., & Titman, S. (1993). Returns to Buying Winners and Selling Losers:
Implications for Stock Market Efficiency. Journal of Finance, 48, 65–91.
Lemmon, M., & Portniaguina, E. (2006). Consumer Confidence and Asset Prices: Some
Empirical Evidence. Review of Financial Studies, 19(4), 1499-1529.
Qiu, L. X., & Welch, I. (2004). Investor Sentiment Measures, NBER Working Paper Series.
Schmeling, M. (2009). Investor sentiment and stock returns: Some international evidence.
Journal of Empirical Finance, 16(3), 394-408.
Schwarz, N. (2002). Situated Cognition and the wisdom of feelings: Cognitive Tuning. In B.
Feldman & P. Salovey (Eds.), The wisdom of feelings (pp. 144 - 166). New York:
Guilford Press.
20
Table 1 Sample Description
No. Country
Stock Exchange
Abbr.
No. of Stocks
1
Bangladesh
Dhaka Stock Exchange
DSE
351
2
China
Shanghai Stock Exchange
SSE
948
3
Hong Kong
Hong Kong Stock Exchange
HKEX
1505
4
India
Bombay Stock Exchange
BSE
3101
5
Indonesia
Indonesia Stock Exchange
IDX
498
6
Japan
Tokyo Stock Exchange
TSE
2913
7
Malaysia
Bursa Malaysia
MYX
1052
8
Pakistan
Karachi Stock Exchange
KSE
447
9
Philippines
Philippine Stock Exchange
PSE
274
10
Singapore
Singapore Exchange
SGX
775
11
South Korea
Korea Exchange
KRX
995
12
Taiwan
Taiwan Stock Exchange
TSEC
926
13
Thailand
Stock Exchange of Thailand
SET
675
Total Sample
14,460
21
Table 2 Average monthly returns for the momentum strategy (%)
Momentum
Bangladesh
Winner
Loser
Momentum
Return (%)
2.829***
1.358*
1.470**
t-stat
(3.76)
(1.72)
(2.27)
China
Return (%)
1.041
0.486
0.555
t-stat
(1.36)
(0.56)
(1.50)
Hong Kong
Return (%)
1.301*
0.320
0.981**
t-stat
(1.82)
(0.35)
(2.06)
India
Return (%)
2.666***
3.226***
-0.560
t-stat
(2.97)
(2.78)
(-0.91)
Indonesia
Return (%)
1.639***
1.363*
0.276
t-stat
(3.03)
(1.90)
(0.55)
Japan
Return (%)
0.156
0.014
0.142
t-stat
(0.32)
(0.02)
(0.43)
Malaysia
Return (%)
0.493
-0.260
0.753*
t-stat
(1.12)
(-0.37)
(1.71)
Pakistan
Return (%)
1.910***
1.298*
0.611
t-stat
(3.21)
(1.72)
(1.08)
Philippines
Return (%)
1.503**
2.442***
-0.939
t-stat
(2.39)
(2.87)
(-1.51)
Singapore
Return (%)
0.964*
0.067
0.897
t-stat
(1.66)
(0.08)
(1.65)
South Korea
Return (%)
1.682**
0.552
1.130**
(2.32)
(0.65)
(2.40)
Taiwan
Return (%)
0.573
0.460
0.113
t-stat
(0.74)
(0.49)
(0.24)
Thailand
Return (%)
1.932***
1.347*
0.585
t-stat
(3.65)
(1.79)
(1.13)
Average
Return (%)
1.438***
0.975***
0.463**
t-stat
(6.47)
(3.49)
(2.51)
*, **,*** represent statistical significance at 10, 5, and 1 percent level respectively.
22
Table 3 Local Investor Sentiment and Momentum Strategy
Country
China
Hong Kong
India
Indonesia
Japan
Malaysia
Philippines
South Korea
Taiwan
Thailand
Average
Winner
-0.074
(-0.06)
3.172*
(1.85)
-5.138
(-1.13)
2.728**
(2.49)
1.182
(1.46)
-3.656*
(-2.11)
1.335
(0.86)
1.354
(0.93)
-4.939
(-1.41)
2.145*
(1.77)
-0.189
(-0.19)
Optimistic
Loser
-0.600
(-0.40)
3.066*
(1.95)
-5.439
(-1.38)
1.263
(1.17)
-0.098
(-0.11)
-5.830**
(-2.66)
2.063
(1.56)
-1.245
(-0.86)
-5.373
(-1.55)
1.640
(1.12)
-1.055
(-1.00)
Mom
0.526
(1.09)
0.106
(0.17)
0.301
(0.27)
1.465**
(2.46)
1.280***
(2.74)
2.174
(1.43)
-0.728
(-0.66)
2.598***
(3.36)
0.434
(0.26)
0.505
(0.55)
0.866**
(2.72)
Winner
1.816
(1.62)
-0.269
(-0.29)
0.460
(0.27)
1.619**
(2.17)
-0.648
(-0.90)
0.849*
(1.74)
2.762**
(2.33)
1.312
(1.44)
0.667
(0.57)
1.368**
(2.11)
0.994**
(3.12)
Mild
Loser
0.983
(0.75)
-1.997*
(-1.81)
3.162
(0.70)
2.035*
(1.98)
-1.203*
(-1.69)
-0.127
(-0.17)
2.991**
(2.04)
0.105
(0.10)
-1.281
(-0.95)
0.392
(0.41)
0.506
(0.90)
*, **,*** represent statistical significance at 10, 5, and 1 percent level respectively.
23
Mom
0.833
(1.44)
1.728***
(3.64)
-2.701
(-0.81)
-0.416
(-0.56)
0.555*
(1.68)
0.976*
(1.97)
-0.229
(-0.22)
1.207**
(2.22)
1.947*
(1.94)
0.976
(1.50)
0.488
(1.15)
Winner
0.152
(0.10)
3.116**
(2.38)
4.444
(1.88)
0.758
(0.68)
1.125
(1.11)
1.379
(1.47)
2.100
(1.47)
4.234**
(2.26)
1.644
(0.52)
4.918***
(4.42)
2.387***
(4.49)
Pessimistic
Loser
0.243
(0.16)
3.013
(1.22)
3.121
(1.53)
-0.101
(-0.06)
3.725*
(2.02)
2.157
(1.20)
1.742
(1.55)
5.608*
(2.06)
1.784
(0.65)
6.535***
(3.62)
2.783***
(4.16)
Mom
-0.091
(-0.14)
0.103
(0.06)
1.323
(0.79)
0.859
(0.82)
-2.599**
(-2.21)
-0.779
(-0.70)
0.358
(0.44)
-1.374
(-0.74)
-0.140
(-0.22)
-1.618
(-0.90)
-0.396
(-1.04)
Table 4 Global Investor Sentiment and Momentum Strategy
Country
Bangladesh
China
Hong Kong
India
Indonesia
Japan
Malaysia
Pakistan
Philippines
Singapore
South Korea
Taiwan
Thailand
Average
Optimistic
Winner
Loser
2.603 **
0.036
(2.19)
(0.03)
1.556
0.269
(1.00)
(0.16)
2.132
-0.772
(1.04)
(-0.32)
1.833 *
1.583
(0.77)
(1.94)
1.073
-0.753
(0.72)
(-0.41)
-0.353
-1.390
(-0.35)
(-1.06)
-0.566
-1.536
(-0.43)
(-0.62)
0.697
2.433 *
(1.97)
(0.57)
1.571
0.849
(0.87)
(0.34)
0.369
-1.537
(0.25)
(-0.77)
1.702
3.436
(1.58)
(0.57)
-0.933
-1.478
(-0.42)
(-0.47)
2.561
2.807 **
(2.53)
(1.00)
1.382 ***
0.018
(3.60)
(0.05)
Mom
2.568 **
(2.46)
1.288 **
(2.74)
2.904 **
(2.55)
0.249
(0.15)
1.826
(1.59)
1.037
(1.47)
0.970
(0.55)
1.736 *
(2.02)
0.723
(0.49)
1.906 *
(1.98)
1.734
(1.24)
0.545
(0.33)
0.246
(0.11)
1.364 ***
(5.86)
Winner
2.560 ***
(3.33)
0.375
(0.34)
0.872
(0.94)
3.359 **
(2.16)
1.455 *
(1.92)
0.407
(0.57)
0.413
(0.71)
2.944 ***
(3.38)
1.239
(1.35)
1.269
(1.49)
1.161
(1.16)
0.614
(0.58)
1.537 *
(1.92)
1.400 ***
(5.14)
Mild
Loser
1.010
(0.92)
-0.644
(-0.51)
-0.497
(-0.46)
4.199 **
(2.30)
1.015
(1.16)
-0.151
(-0.19)
-0.855
(-1.09)
2.595 **
(2.55)
2.253 **
(2.09)
-0.329
(-0.29)
-0.524
(-0.52)
0.529
(0.45)
0.255
(0.28)
0.681
(1.62)
Mom
1.549
(1.57)
1.019 *
(1.80)
1.369 ***
(2.98)
-0.840
(-1.13)
0.440
(0.87)
0.558
(1.41)
1.269 ***
(3.06)
0.349
(0.63)
-1.013
(-1.25)
1.598 **
(2.47)
1.686 ***
(2.93)
0.084
(0.15)
1.281 **
(2.60)
0.719 **
(2.90)
*, **,*** represent statistical significance at 10, 5, and 1 percent level respectively
24
Pessimistic
Winner
Loser
3.502
2.880 *
(1.64)
(1.73)
2.031
2.856 *
(1.43)
(1.82)
1.622
2.620
(1.25)
(1.31)
1.826 **
2.344
(2.10)
(1.38)
2.358 ***
3.384 **
(2.79)
(2.24)
-0.018
1.228
(-0.02)
(1.04)
1.319 *
1.723
(1.85)
(1.48)
-0.463
-0.881
(-0.49)
(-0.54)
1.980 **
3.823 **
(2.45)
(2.41)
0.736
1.861
(0.86)
(1.05)
1.601
1.949
(1.54)
(1.46)
1.442
1.546
(1.22)
(1.00)
2.161 **
2.734 **
(2.60)
(2.30)
1.546 ***
2.159 ***
(5.45)
(6.61)
Mom
0.622
(0.56)
-0.825
(-1.33)
-0.998
(-0.82)
-0.518
(-0.39)
-1.026
(-0.79)
-1.246
(-1.66)
-0.403
(-0.51)
0.418
(0.25)
-1.843
(-1.43)
-1.125
(-0.85)
-0.347
(-0.39)
-0.104
(-0.14)
-0.574
(-0.67)
-0.613 ***
(-3.26)
Table 5 Momentum Returns Using Alternative Sentiment Classification
Country
Bangladesh
China
Hong Kong
India
Indonesia
Japan
Malaysia
Pakistan
Philippines
Singapore
South Korea
Taiwan
Thailand
Local Sentiment
Optimistic Pessimistic
N/A
N/A
0.668
(1.53)
0.343
(0.68)
2.288
(1.68)
1.105 **
(2.06)
1.025 **
(2.61)
1.560
(1.59)
N/A
-0.761
(-0.73)
N/A
1.749 **
(2.53)
2.074
(1.18)
1.174
(1.67)
1.097
(1.60)
0.879
(0.78)
0.786
(0.68)
0.945
(1.19)
-1.239
(-1.57)
0.149
(0.16)
N/A
0.169
(0.17)
N/A
0.316
(0.26)
-0.258
(-0.54)
-0.471
(-0.49)
Global Sentiment
Optimistic Pessimistic
2.101 **
1.296
(2.49)
(1.38)
0.257
-0.090
(0.41)
(-0.15)
1.446
-0.354
(1.59)
(-0.34)
0.107
-0.672
(0.08)
(-0.56)
1.515 *
-0.312
(1.76)
(-0.28)
1.341 **
-1.176 *
(2.40)
(-1.72)
1.493
-0.294
(1.22)
(-0.42)
1.131
0.273
(1.66)
(0.20)
0.760
-1.457
(0.71)
(-1.30)
1.667 **
-1.081
(2.28)
(-0.87)
1.769 *
0.652
(1.73)
(0.84)
0.500
-0.140
(0.42)
(-0.22)
0.727
-0.362
(0.48)
(-0.50)
*, **,*** represent statistical significance at 10, 5, and 1 percent level respectively
25
Table 6 Momentum Portfolio Returns Using Alternative Sentiment Proxies
Country
Bangladesh
China
Hong Kong
India
Indonesia
Japan
Malaysia
Pakistan
Philippines
Singapore
South Korea
Taiwan
Thailand
University of Michigan Sentiment Index
Optimistic
Mild
Pessimistic
4.949 **
0.910
0.710
(2.52)
(1.06)
(0.67)
2.153 *
0.772
-0.697
(2.06)
(1.55)
(-1.16)
1.785
1.620 ***
-0.655
(1.45)
(3.45)
(-0.55)
-1.790
-0.492
-0.045
(-0.63)
(-0.03)
(-1.19)
1.573 *
0.413
-0.665
(1.88)
(0.74)
(-0.52)
0.402
0.659
-0.977
(0.88)
(1.56)
(-1.32)
3.401 **
0.453
-0.069
(2.74)
(0.81)
(-0.09)
0.514
0.403
1.358
(1.72)
(0.89)
(0.26)
-0.298
-0.945
-1.265
(-0.14)
(-1.34)
(-0.98)
2.834 ***
1.220 *
-0.734
(3.00)
(1.90)
(-0.57)
1.973 ***
-0.223
0.657
(0.41)
(3.50)
(-0.26)
0.031
0.248
-0.102
(0.02)
(0.43)
(-0.14)
1.167 **
-0.333
0.228
(0.09)
(2.46)
(-0.40)
Baker and Wurgler (2006) Composite Index
Optimistic
Mild
Pessimistic
1.135
0.283
4.860 ***
(1.31)
(0.28)
(3.64)
0.515
0.630
1.023
(1.13)
(1.02)
(1.31)
3.118 ***
0.143
0.628
(3.28)
(0.19)
(0.78)
-1.271
-0.464
-0.833
(-0.86)
(-0.49)
(-0.73)
1.860 **
-0.669
0.647
(2.05)
(-0.82)
(0.77)
0.278
0.417
-0.500
(0.41)
(0.85)
(-0.65)
0.869
0.898
0.335
(0.60)
(1.58)
(0.53)
0.545
0.771
0.111
(0.54)
(0.81)
(0.12)
-0.554
-0.722
-1.997 *
(-0.38)
(-0.75)
(-1.84)
1.741
1.144
-0.910
(1.61)
(1.41)
(-0.71)
1.355
0.612
2.174 ***
(0.98)
(1.00)
(3.12)
-1.897
0.624
0.017
(-1.23)
(1.08)
(0.03)
-0.100
1.063 *
0.120
(-0.06)
(1.74)
(0.15)
*, **,*** represent statistical significance at 10, 5, and 1 percent level respectively.
26
Table 7 Firm Size, Local Investor Sentiment and Momentum Portfolio Returns
Country
China
Hong Kong
India
Indonesia
Japan
Malaysia
Philippines
South Korea
Taiwan
Thailand
Large
0.759
(1.38)
1.623 *
(1.94)
0.743
(0.64)
1.299
(1.59)
0.850
(1.55)
1.419
(1.20)
2.021
(1.45)
1.983 *
(1.73)
-0.832
(-0.42)
1.753
(1.53)
Optimistic
Medium
0.500
(1.20)
1.527 **
(2.21)
0.627
(0.39)
2.214 **
(2.34)
1.020 **
(2.49)
2.839 *
(1.98)
-0.694
(-0.61)
2.476 ***
(2.89)
1.812
(0.74)
-0.223
(-0.19)
Small
0.321
(0.62)
-0.039
(-0.04)
1.567
(1.18)
1.246
(1.45)
1.081 *
(1.79)
1.730
(1.43)
-1.411
(-0.94)
1.573
(1.43)
1.091
(0.80)
0.207
(0.18)
Large
0.833
(1.30)
1.824 ***
(3.29)
-2.118
(-0.56)
0.144
(0.14)
0.101
(0.23)
0.942 **
(2.08)
0.504
(0.48)
1.199
(1.55)
2.565 *
(2.04)
0.978
(1.33)
Mild
Medium
0.594
(1.24)
1.970 ***
(3.75)
-2.449
(-0.70)
0.168
(0.20)
0.772 **
(2.22)
1.037 **
(2.27)
-0.690
(-0.50)
1.229 **
(2.53)
1.538 *
(1.96)
1.463 **
(2.18)
*, **,*** represent statistical significance at 10, 5, and 1 percent level respectively.
27
Small
0.111
(0.30)
1.653 ***
(2.70)
-4.057
(-1.48)
-0.667
(-0.73)
0.981 ***
(2.73)
0.251
(0.40)
0.469
(0.33)
0.938 *
(1.83)
1.557
(1.68)
1.103 *
(1.72)
Large
0.158
(0.16)
-1.514
(-0.97)
0.901
(1.69)
0.130
(0.07)
-2.453 **
(-2.07)
-0.870
(-0.56)
-0.258
(-0.19)
-0.353
(-0.17)
-1.333
(-1.40)
-2.431
(-0.93)
Pessimistic
Medium
0.089
(0.15)
2.183
(1.34)
3.938 **
(3.59)
1.463
(1.03)
-2.351 *
(-1.86)
-0.184
(-0.14)
2.010
(1.51)
-1.906
(-1.15)
0.790
(1.40)
0.160
(0.10)
Small
-0.828 **
(-2.06)
-1.365
(-0.77)
0.199
(0.15)
1.618
(0.99)
-2.223 *
(-1.80)
0.052
(0.05)
-1.922
(-1.49)
-1.684
(-0.79)
-0.378
(-0.41)
-2.477
(-1.55)
Table 8 Firm Size, Global Investor Sentiment and Momentum Portfolio Returns
Country
Bangladesh
China
Hong Kong
India
Indonesia
Japan
Malaysia
Pakistan
Philippines
Singapore
South Korea
Taiwan
Thailand
Optimistic
Large
Medium
1.471
1.102
(1.00)
(1.32)
2.256 ***
0.977 **
(3.23)
(2.37)
3.074 **
4.335 ***
(2.36)
(3.39)
1.121
0.108
(0.06)
(0.44)
2.046
4.165 **
(0.97)
(2.68)
0.656
1.367 *
(0.80)
(2.05)
0.928
1.750
(0.61)
(1.07)
2.608 ***
1.380
(1.04)
(2.96)
-0.977
1.475
(-0.42)
(0.89)
1.225
2.952 **
(1.09)
(2.32)
1.009
3.209
(1.37)
(0.76)
1.156
0.851
(0.72)
(0.46)
0.386
0.102
(0.04)
(0.18)
Small
0.812
(0.84)
-0.449
(-0.75)
2.576 *
(2.03)
-0.011
(-0.01)
-0.002
(0.00)
0.597
(0.75)
0.561
(0.24)
0.611
(0.48)
1.406
(0.65)
2.554 *
(2.03)
-0.640
(-0.50)
-0.028
(-0.01)
1.609
(0.89)
Large
1.739 **
(2.42)
0.862
(1.50)
1.468 ***
(3.09)
1.188 *
(1.75)
0.750
(0.96)
0.211
(0.47)
1.233 ***
(2.89)
0.624
(0.70)
-0.222
(-0.22)
1.482 ***
(2.70)
1.390 **
(2.04)
1.126 *
(1.81)
1.964 ***
(3.22)
Mild
Medium
2.210 ***
(2.85)
0.765
(1.64)
2.583 ***
(5.36)
0.310
(0.40)
0.221
(0.29)
0.559
(1.34)
1.345 ***
(3.32)
0.937 *
(1.76)
-1.397
(-1.30)
2.206 ***
(4.14)
1.690 ***
(3.41)
0.307
(0.55)
1.349 **
(2.19)
*, **,*** represent statistical significance at 10, 5, and 1 percent level respectively.
28
Small
-0.539
(-0.51)
0.697 **
(2.03)
0.544
(0.91)
-2.263 **
(-2.61)
1.094
(1.50)
1.063 **
(2.57)
0.488
(1.06)
-0.890
(-0.98)
-0.801
(-0.77)
1.557 *
(1.81)
1.348 **
(2.09)
0.175
(0.31)
0.942
(1.51)
Large
1.263
(1.10)
-0.747
(-0.78)
-1.155
(-0.89)
0.339
(0.23)
-1.416
(-0.84)
-1.525 *
(-1.87)
-0.759
(-0.74)
-0.442
(-0.26)
-0.161
(-0.13)
-1.073
(-0.66)
-0.648
(-0.61)
-0.100
(-0.12)
-1.128
(-0.98)
Pessimistic
Medium
0.656
(0.55)
-0.481
(-0.83)
-0.934
(-0.83)
0.085
(0.06)
0.172
(0.15)
-1.092
(-1.42)
-0.320
(-0.36)
0.487
(0.26)
-2.073
(-1.27)
-1.474
(-0.91)
-0.157
(-0.18)
0.085
(0.10)
0.478
(0.58)
Small
-0.492
(-0.27)
-1.328 ***
(-3.16)
-0.628
(-0.44)
-1.605
(-1.34)
-1.167
(-0.85)
-1.039
(-1.25)
-0.087
(-0.13)
1.251
(0.65)
-2.788 **
(-2.32)
-1.116
(-1.08)
0.354
(0.39)
0.017
(0.02)
-1.044
(-1.11)
Table 9 Trading Volume, Local Investor Sentiment and Momentum Portfolio Returns
Country
China
Hong Kong
India
Indonesia
Japan
Malaysia
Philippines
South Korea
Taiwan
Thailand
High Vol.
0.853*
(1.74)
1.078
(1.48)
-0.495
(-0.16)
1.207
(1.32)
1.773***
(2.77)
0.982
(0.99)
0.332
(0.15)
2.201*
(2.03)
-0.888
(-0.33)
0.138
(0.11)
Optimistic
Med. Vol.
0.278
(0.52)
-0.069
(-0.11)
1.630
(0.66)
2.741**
(2.70)
1.086**
(2.33)
1.715
(1.50)
-0.851
(-0.61)
2.640***
(2.97)
0.751
(0.47)
1.066
(0.87)
Low Vol.
0.359
(0.60)
-0.493
(-0.57)
-1.458
(-0.95)
1.171*
(1.77)
0.855**
(2.37)
2.060*
(2.09)
-0.114
(-0.07)
3.318***
(3.90)
3.391**
(3.16)
-0.669
(-0.63)
High Vol.
0.728
(1.11)
1.289**
(2.24)
-1.491
(-0.33)
-1.054
(-1.11)
0.394
(0.98)
1.314**
(2.62)
-0.384
(-0.30)
1.140**
(2.06)
2.787*
(2.18)
0.940
(1.27)
Mild
Med. Vol.
0.874
(1.43)
1.982***
(3.94)
-1.391
(-0.58)
0.511
(0.57)
0.764**
(2.38)
1.012**
(2.08)
0.557
(0.46)
1.351**
(2.29)
2.776**
(2.38)
0.854
(1.17)
*, **,*** represent statistical significance at 10, 5, and 1 percent level respectively.
29
Low Vol.
0.435
(0.74)
2.228***
(3.73)
-1.343
(-1.51)
-0.340
(-0.42)
0.906***
(3.13)
0.966*
(1.83)
-0.547
(-0.41)
1.329**
(2.05)
1.335
(1.34)
1.760**
(2.12)
High Vol.
-0.018
(-0.04)
-0.251
(-0.15)
0.373
(0.18)
1.880
(1.38)
-2.112
(-1.69)
-1.483
(-1.00)
0.272
(0.24)
-0.561
(-0.37)
0.570***
(5.64)
-2.179
(-1.22)
Pessimistic
Med. Vol.
-0.507
(-0.85)
0.669
(0.41)
-4.018
(-0.72)
1.373
(0.95)
-3.032**
(-2.56)
-0.483
(-0.38)
0.452
(0.44)
-0.680
(-0.34)
0.805
(0.80)
-1.905
(-0.98)
Low Vol.
0.140
(0.16)
-0.747
(-0.59)
-7.297
(-1.42)
0.171
(0.13)
-2.049*
(-1.87)
0.727
(0.84)
-0.535
(-0.38)
-2.221
(-1.22)
-0.414
(-0.39)
-1.190
(-0.58)
Table 10 Trading Volume, Global Investor Sentiment and Momentum Portfolio Returns
Country
Bangladesh
China
Hong Kong
India
Indonesia
Japan
Malaysia
Pakistan
Philippines
Singapore
South Korea
Taiwan
Thailand
High Vol.
1.930
(1.61)
0.568
(0.81)
1.232
(0.83)
-0.240
(-0.08)
1.885
(1.10)
0.946
(0.98)
0.428
(0.29)
0.521
(0.43)
1.691
(1.04)
1.520
(1.41)
1.059
(0.77)
0.141
(0.08)
0.059
(0.03)
Optimistic
Med. Vol.
3.038 ***
(3.04)
1.292 **
(2.13)
2.917 **
(2.38)
0.937
(0.54)
3.546 *
(2.00)
1.274 *
(1.84)
0.984
(0.60)
2.637 *
(1.80)
-0.756
(-0.50)
3.048 **
(2.66)
2.120
(1.32)
-0.233
(-0.13)
-0.461
(-0.19)
Low Vol.
1.068
(0.85)
1.377 *
(2.07)
3.141 **
(2.37)
-0.101
(-0.13)
2.161
(1.34)
0.874
(1.65)
0.787
(0.45)
1.295
(0.82)
1.466
(0.54)
2.449 **
(2.42)
1.685
(0.85)
1.076
(0.49)
1.855
(0.68)
High Vol.
2.718 ***
(3.14)
1.122 *
(1.76)
1.774 ***
(4.10)
0.865
(0.95)
0.242
(0.30)
0.881 **
(2.04)
1.596 ***
(3.25)
1.064
(1.39)
-2.010
(-1.56)
2.369 ***
(3.89)
1.495 **
(2.48)
0.539
(0.91)
1.111
(1.61)
Mild
Med. Vol.
1.411
(1.62)
0.918
(1.52)
1.542 ***
(3.00)
-0.419
(-0.59)
0.937
(1.33)
0.440
(1.06)
1.113 **
(2.60)
0.816
(1.09)
-0.911
(-0.95)
1.441 *
(1.90)
2.009 ***
(3.25)
0.006
(0.01)
1.659 ***
(2.64)
*, **,*** represent statistical significance at 10, 5, and 1 percent level respectively.
30
Low Vol.
-0.366
(-0.32)
0.612
(1.11)
0.552
(0.99)
-2.128 **
(-2.52)
0.092
(0.14)
0.620
(1.60)
1.447 ***
(3.07)
-1.149 *
(-1.72)
-1.290
(-1.37)
0.893
(1.15)
1.610 **
(2.59)
-0.077
(-0.12)
1.324 **
(2.05)
High Vol.
1.186
(0.84)
-0.483
(-1.00)
-1.065
(-0.80)
0.540
(0.39)
-1.256
(-0.91)
-1.358
(-1.64)
-0.775
(-0.75)
2.324
(1.46)
-0.995
(-0.56)
-0.557
(-0.42)
0.398
(0.46)
0.295
(0.33)
-0.725
(-0.79)
Pessimistic
Med. Vol.
0.486
(0.48)
-1.079 *
(-1.97)
-0.540
(-0.48)
0.382
(0.36)
0.144
(0.10)
-1.219
(-1.68)
-0.011
(-0.01)
1.063
(0.72)
-1.196
(-0.89)
-0.806
(-0.76)
-0.528
(-0.60)
0.693
(0.93)
-0.820
(-0.92)
Low Vol.
-1.107
(-0.89)
-0.813
(-0.95)
0.240
(0.22)
-1.405
(-1.25)
-1.215
(-1.08)
-0.577
(-0.91)
0.306
(0.59)
0.055
(0.03)
-1.929
(-1.30)
-0.399
(-0.33)
0.217
(0.25)
-0.111
(-0.13)
-0.585
(-0.63)