“Various moving average convergence divergence trading strategies:

a comparison”

AUTHORS Nguyen Hoang Hung

Nguyen Hoang Hung (2016). Various moving average convergence

divergence trading strategies: a comparison. Investment Management
ARTICLE INFO
and Financial Innovations (open-access), 13(2-2). doi:10.21511/imfi.13(2-
2).2016.11

DOI http://dx.doi.org/10.21511/imfi.13(2-2).2016.11

JOURNAL "Investment Management and Financial Innovations (open-access)"

NUMBER OF REFERENCES NUMBER OF FIGURES NUMBER OF TABLES

0 0 0

© The author(s) 2017. This publication is an open access article.

businessperspectives.org
 Investment Management and Financial Innovations, Volume 13, Issue 2, 2016

Nguyen Hoang Hung (Vietnam)

Various moving average convergence divergence trading

strategies: a comparison
Abstract
Some studies published recently (Dejan Eric, 2009; R. Rosillo, 2013; Terence Tai-Leung Chong, 2008; Ülkü and Prodan,
2013) uncover that moving average convergence divergence (MACD) trading rules have predictive ability in many countries.
The MACD trading strategies applied by these papers to execute the trading signals are various. This study analyzes the
performance of a MACD trading strategy (MACD-4 in the current study), which is applied popularly by practitioners, but
was not tested by prior academicians. Furthermore, the author compares the performance of each of the strategies on a group
of markets to identify the best one. Before considering the costs, the author finds that the MACD-4 trading strategy has
predictive ability. The best performance is MACD strategy applied by Terence Tai-Leung Chong (2008). This strategy is also
the most effective one if it is applied in a high trading cost environmentm because the numbers of trades created are the
lowest. Especially, the strategy applied by R. Rosillo (2013) is unpredictable in the selected samples.
Keywords: moving average convergence divergence, technical trading rules, excess returns, transaction costs,
profitability, MACD trading strategy.
JEL Classification: G15, G15, G17.

Introduction ” have been conducted. In the field of MACD technical

rules, we aware a few papers which the highlights of
Moving average convergence divergence (MACD) is
these are the study of Terence Tai-Leung Chong
one of the most popular technical indicators. It was
(2008) on FT30; study of Dejan Eric (2009) on the
invented by Gerald Appel in the 1970’s. MACD is
financial market of The Republic of Serbia; study of
calculated using two exponential moving averages and
Terence Tai-Leung Chong (2010) on the BRIC
its value equals the value of the shorter period (faster)
countries; Ülkü and Prodan (2013) on a group of 30
EMA less the value of the longer period (slower)
countries; R. Rosillo (2013) on Spanish market
EMA. Therefore, the MACD line itself always
companies. The MACD trading strategies which these
represents the distance between the two moving
papers apply to execute the trading signals are various.
averages. When MACD line crosses the zero line, it is
exactly at the same time when the faster EMA crosses Table 1 records the various MACD trading strategies
the slower EMA. On a typical chart, the MACD shows conducted by researchers on stock markets so far.
two lines (the MACD line and the signal line), as well From the Table, there have been definitely three
as the zero line. They cross each other often enough to strategies for applying MACD trading rules to
give traders plenty of options for building examine trading signals. In the study of Terence Tai-
various MACD crossover trend following strategies. A Leung Chong (2008), the MACD trading signals are
simplest strategy is trading on MACD line crossing the examined when the MACD line crosses the zero line.
zero line that a buy signal is generated when the For saving space, we call this strategy as MACD-1
MACD crosses above the zero line. When the MACD strategy in the present paper. Another way, in the
crosses below the zero line, then, a sell signal is studies of Dejan Eric (2009), Terence Tai-Leung
generated. MACD line crossing the signal line is Chong (2010) and Ülkü and Prodan (2013), the
another popular strategy from which a buy signal is MACD trading signals are generated when MACD
generated when the MACD crosses above the signal line crosses its signal line, we call this method as
line. Similarly, when the MACD line crosses below MACD-2 strategy. Specially, in study of R. Rosillo
the signal line a sell signal is generated. An alternative (2013), the trading signals are occurred when both
is using the signal line instead of the MACD line and MACD and signal lines cross zero line, besides
enter trades when it crosses the zero line. It is as MACD line crosses signal line, we call this strategy as
moothed (lagged) version of the MACD line, as it is MACD-3 strategy. By my contacts to practitioners, as
calculated as an EMA of the MACD. well as my experience in applying MACD rules on
stock markets, the MACD-3 strategy is rarely to be
Following the study of (William Brock, 1992) which
used by traders, because applying this strategy does
tested several moving average lengths and found them
not create profit for investors, and this system almost
useful in predicting stock prices, a numbers of studies
makes investors loss money in the real markets. By
on testing the profitability of technical trading rules
this system, buy signal is generated when the MACD
and signal line cross below the zero line, besides signal
” Nguyen Hoang Hung, 2016.
line is greater than the MACD line. Similarly, sell
Nguyen Hoang Hung, Ph.D., Faculty of Finance and Banking, Industrial signal is generated when both the MACD and signal
University of Ho Chi Minh City, Vietnam. lines cross above the zero line, besides, signal line is

363
Investment Management and Financial Innovations, Volume 13, Issue 2, 2016

less than the MACD line. By its definition, MACD is selections”. Any successful results may be got by
calculated using two exponential moving averages and chance because of data snooping problem. Bootstrap
its value equals the value of the faster period EMA less reality check methodology, which is applied by
the value of the slower period EMA. Therefore, the (Sullivan, 1999) has been used to quantify the effect of
MACD line itself always represents the distance data snooping problem. However, an important issue
between the two moving averages. When MACD line of this method is the construction of “full universe” of
crosses the zero line, it is exactly at the same time technical trading rules. Sullivan (1999) assumes that
when the faster EMA crosses the slower EMA. rules from five simple technical trading rules represent
Besides, signal line is a lagging of MACD, as it is the full set of technical trading rules. However, there
calculated as an EMA of the MACD line. may be a huge number of trading rules that may not
Consequently, at the time when MACD line crosses include in their full set of trading rules. Thus, if a set of
below the zero line and the signal line is greater than trading rules testing is a subset of a “full universe” of
the MACD, trading signal should be a sell signal1, not technical trading rules, the applying of this method
a buy one which is determined in MACD-3 strategy. may still be subject to result bias. Moreover, the
Based on this argument, we recommend the MACD-4 present paper does not test a universe of technical
strategy which buy signal is generated when both the trading rules, thus, the bootstrap reality check is not
MACD and signal line cross above the zero line, relevant in the current paper.
besides, signal line is less than the MACD. Sell signal
is generated when the MACD and signal line cross Using the most popular trading rule on the same data
below the zero line, besides signal line is greater than set is also a kind of data snooping problem. When
the MACD. Actually, MACD-4 strategy is an opposite researchers test a popular rule on a period during
way of MACD-3 one, from which the “buy signal” in which it became popular, he or she is actually using
MACD-3 strategy becomes “sell signal” in MACD-4 the past data twice (Sullivan, 1999). MACD is
strategy and the “sell signal” in MACD-3 strategy invented by Gerald Appel during the late 1970’s. It
becomes the “buy signal” in MACD-4 strategy. The has become one of the most popular technical tools,
MACD-4 strategy have not found in academic thus, testing MACD ought to be kind of data
researches so far, even though this strategy has been snooping bias in the current research. Ülkü and
used popularly by traders. Prodan (2013) give us a way to solve this problem,
In our paper, we examine the profitability of MACD-4 just test the trading rules on the samples that the
strategy and, then, compare the profitability of trading rules had become popular before the
each of the four MACD trading strategies (MACD- samples exist. They state that “If the sample covers
1, MACD-2, MACD-3 and MACD-4 strategies) on the time period over which a technical rule has
a group of markets to confirm the best one. Before become popular, the test is subject to the
extracting trading costs, we find that the MACD-1 survivorship bias. However, if the researcher
strategy has the highest powerful profitability, employs an out-of-sample test of a technical rule
following is MACD-4 strategy and the lowest which was already popular before the beginning of
predictability is MACD-2 strategy. Specially, the his/her sample period (i.e., become popular in the
figures show that returns created by MACD-3 past), then the study is free of subtle form of data
strategy are unprofitable in both base and new snooping bias”. The bias sample and new fresh
fresh sample markets. This analysis makes three sample periods in the present study are from 2001 to
contributions to the literature: first, giving 2012, after the MACD trading rule become popular.
evidence to refuse the using of the MACD strategy In the present paper, we, first, replicate the same
(MACD-3 strategy) applied in the study of MACD trading strategies which are MACD-1,
R. Rosillo (2013), second, examining the MACD-2, MACD-3 strategies as the original works of
profitability of MACD-4 strategy which has not Terence Tai-Leung Chong (2008), Ülkü and Prodan
been noticed by previous researchers, third, (2013) and R. Rosillo (2013), respectively, as well as
comparing the profitability of each of the four our MACD-4 trading strategy on the extending
MACD strategies to confirm the best one.
samples of Japanese, German and Russian markets
Data snooping problem is one of the main interests of which are the base sample in the study of Ülkü and
academicians in technical analysis. According to Prodan (2013)2. Second, the frameworks are
Sullivan (1999), “data snooping happens when a set of duplicated on Shanghai and Vietnamese markets
data is used more than once for purposes of model which are not examined in the study of2 Ülkü and

2
Their original samples include 30 national indexes ranged from frontier
1
The moving average strategy defined by William Brock (1992): “ If to developed markets. However, only RTS, Nikkei 225, Dax markets are
the short moving average is above (below) the long, the day is classified chosen in the present study because the MACD trading rule has been
as a buy (sell)”. proven to have high profitability in these markets in their study.

364
 Investment Management and Financial Innovations, Volume 13, Issue 2, 2016

Prodan (2013)3, over the corresponding period. This predictability on the two markets, even though it is
step is also a procedure for a double checking of our not statistically significant. The choosing of RTS in
results. To minimize the data snooping problem, our research is reliable and more meaningful. The
some authors choose to replicate the evaluation on RTS index is proved to have highly statistically
the more recent sample periods robust to the significant profitability with MACD trading rule.
survivorship bias. In the current study, however, 1 Our new fresh data samples are SSE and HOSE
the similar procedure may be not meaningful, indexes. The SSE is an emerging market, whereas
because the time series data are not long enough. HOSE is a frontier one. The emerging and frontier
Moreover, there are many unusual events which markets have been proven to have high profitable
lead to results bias happened in the January 2013- power with technical analysis. Some of these studies
January 2015 period, such as a heavy crisis in are Gunasekarage and Power (2001), Bessembinder
Russia, a monetary policy intervention in Japan4. and Chan (1995), Ming-Ming and Siok-Hwa (2006),
Those are reasons to explain why we choose to Yu, Nartea, Gan and Yao (2013), Nguyen Hoang
replicate 2 the framework on the new fresh data Hung (2013). We source Nikkei 225, Dax, RTS and
which are SSE and HOSE samples, instead of on the SSE series data from Yahoo.com/finance whereas
more recent out-of data samples. HOSE series data are downloaded from
www.vietstock.vn, one of the most popular financial
1. Data and technical trading rules services in Vietnam. The daily returns are calculated
1.1. Data. Our base samples include developed, as log differences of the samples.
emerging markets whereas our new fresh samples 1.2. Technical trading rules. MACD is computed
include emerging and frontier markets. The data are by subtracting a longer exponential moving average
adjusted closing prices of Japanese Nikkei 225 (EMA) from a shorter one. EMA is calculated as
index (Nikkei 225), German Dax index (Dax), follows:
Russian RTS index (RTS), Shanghai SSE index
2
(SSE) and Ho Chi Minh stock index (HOSE), over EMA n Pt  EMAt 1  EMAt 1 , (1)
the period of 2001-2012. Our base sample is longer
t
n 1 
than the original one in paper of Numan Ulkü where where Pt is the closing level of the national index on
their base period is from 2002 to 2011. Our base day t, and n is the number of periods for calculating
samples have to meet at least two criteria: one is to EMA. The initial EMA is the n – day simple
try to keep as the same as the original period in the moving average of the series. Then, MACD is
original study to attempt to link our results to those computed as:
of their study, another is to ensure that the chosen
period sample is as long as possible5. In the present MACDt EMA s t  EMA l t . (2)
paper, our main target is not testing the 3profitability
Further, an EMA of the MACD line itself is
of MACD trading rule. Rather, we compare the
computed to generate signals, and is called signal
performances of each trading strategy among the
line. Accordingly, an MACD rule can be described
four MACD ones to confirm the most profitable
as MACD(s, l, k) where s and l are the lag lengths of
one. Thus, the sample chosen must meet important
short (fast) and long (slow) EMA, respectively, and
criteria, from which the technical trading rule used
k is the lag length of the signal line. We use the
to compare must be proven to have predictability on most popular MACD parameters among
that sample. The Nikkei 250 and DAX are practitioners, which is also the default parameter for
developed markets where the markets are widely technical analysis software supplied by most data
considered to be more efficient and less subject to vendors: MACD (12, 26, 9).
problems such as political instability and
government intervention than others. Importantly, Buy and sell signals are determined as follows:
the MACD trading rule to be proven to have MACD-1 strategy: buy signal is generated when the
MACD line crosses above the zero line. Sell signal
3
is generated when the MACD line crosses below the
Shanghai stock index (SSE) is an emerging market whereas Ho Chi
Minh stock index (HOSE) is a frontier one. The emerging and frontier
zero line.
markets have been approved to have high predictability with technical
trading rule. In the study of (Numan Ulkü, Eugeniu Prodan, 2013),
MACD-2 strategy: buy signal is generated when the
MACD trading rule has also been showed to have significant MACD line crosses above the signal line. Sell signal
profitability in emerging and frontier markets. is generated when MACD line crosses below the
4
In order to detect the impact, if any, of the financial crisis in 2008 on
the result, (Fang, J., B. Jacobsen, and Y. Qin, 2014) remove the crisis
signal line.
period in their research.
5
Lakonishok, J. and Smidt, S. (1988) prescribe long and new data series
MACD-3 strategy: buy signal is generated when
as the best remedy against data snooping, noise and ‘boredom’ both the MACD line and signal line cross below the
(selection bias). zero line, besides signal line is greater than the

365
Investment Management and Financial Innovations, Volume 13, Issue 2, 2016

MACD line. Sell signal is generated when both the confirm that the profitability of MACD-2 trading
MACD line and signal line cross above the zero strategy in our base sample has the same magnitude
line, besides signal line is less than the MACD line. with that in the study of Ülkü and Prodan (2013)6. It
is necessary to remind that MACD-2 strategy is
MACD-4 strategy: buy signal is generated when
exactly the MACD trading strategy applied in the
both the MACD line and signal line cross above the
study of Ülkü and Prodan (2013). This confirmation
zero line, besides signal line is less than the MACD
is important, because it guarantees that our results are
line. Sell signal is generated when both the MACD
relevant to previous researches and, thus, it allows us
line and signal line cross below the zero line,
to compare the profitability of each strategy. The
besides signal line is greater than the MACD line.
figures in the table show that the returns of MACD-3
For the four strategies, buy and sell signals are strategy are unprofitable in all sample markets. The
assumed to be executed at the closing level of the buy minus sell returns are significantly negative in
day on which a cross-over takes place. Hence, RTS market whereas they are negative in Nikkei 225
returns to a signal position start to accrue the next and Dax markets. Among the three profitable MACD
day. In the simulation, we assume short positions strategies, the highest profitable strategy is MACD-1
upon sell signals instead of investing in the risk-free one, following is MACD-4 strategy and the least
asset. By construction, the technical trader has profitable strategy is the MACD-2 one. Indeed, the
always either a long or a short position in the market mean daily buy minus sell returns produced from
(i.e., neutral position is not permitted). Alternative MACD-1 strategy is 0.00058 percent on average
constructions, such as investing in risk-free asset whereas the MACD-2 and MACD-4 strategies
upon sell signals, lead to small variation in results produce the same mean return of 0.00038 percent.
that would not alter our results. In all the MACD The profitability MACD-1 strategy is 20 percent
trading strategies, traders are out of markets when higher than those of MACD-2 or MACD-4 ones. The
MACD line or signal line equals zero. numbers of trades generated from MACD-1 strategy
is 89 whereas they are 229 and 149 for MACD-2 and
2. Empirical results
MACD-4 strategies, on average, respectively. Thus,
We use t-statistic for the differences of mean daily after considering the trade numbers, the performance
returns from zero, obtain from bootstrap simulation of MACD-4 strategy is better than that of the
by reshuffling the return series with 500 replications MACD-2 strategy, and the highest performance is
which is the same test as in study of Ülkü and still MACD-1 strategy. Furthermore, the lowest
Prodan (2013). If the two technical trading rules numbers of trades created by MACD-1 strategy
have similar returns, we prefer the strategy with the suggest traders that this trading strategy is the most
least trade numbers. effective if it is applied in a high trading cost
environment. In terms of excess returns, the results
Table 3 shows the signs of predictability of MACD-4
have the same order. On average for the three
strategy. Disregarding trading costs, except for the
markets, the mean daily mean excess return created
sell return where the profit is not statistically
from MACD-1 strategy is 0.00033 percent, whereas
significantly, the buy and buy minus sell results are
they are 0.00014 percent in MACD-2 and 0.00013
statistically significantly profitable at 5% level in percent in MACD-4 strategies. Although the mean
RTS market. The results are profitable, but not daily excess return of MACD-4 strategy is 0.00013,
significant in Nikkei 225 and Dax markets. In terms slightly lower than that of MACD-2 strategywith
of excess return, applying this strategy can create 0.00014 percent, the mean daily excess returns of
positive excess returns for investors. The mean daily MACD-4 strategy is still higher than that of the
excess returns are 0.00002 percent, 0.00012 percent MACD-2 one, because the trade numbers generated
and 0.00025 percent in RTS, Dax and Nikkei 225, from the MACD-2 strategy is far bigger than those
respectively, on average for the three markets. from MACD-4 strategy. 1
Besides, the highest unconditional daily return is
happened in RTS with 0.00077 percent, whereas they Table 5 describes the results of the four MACD
are 0.00006 percent and -0.00009 percent in Dax and trading strategies on the Shanghai stock exchange
Nikkei 225, respectively. The results reveal that the and Ho Chi Minh stock exchange. In general, on
markets are more volatile in the Dax and Nikkei 225 average for the two markets, the results in the new
than that in RTS during the sample period. fresh sample confirm the above findings in the base
sample results. Both the mean daily buy minus sell
Table 4 displays the results of the four MACD returns and mean daily excess returns produced by
strategies. Not taking transaction costs into account.
The buy minus sell return of MACD-2 strategy is
6
significantly positive at 5% level in RTS market, Omitting transaction costs, mean daily buy minus sell return of MACD
trading strategy is significantly positive at 10% level in RTS market
whereas the positive returns are not significant in whereas the returns are positive but not significant in Nikkei 225 and
Nikkei 225 and Dax in the base sample. The results Dax markets.

366
 Investment Management and Financial Innovations, Volume 13, Issue 2, 2016

MACD-1 strategy are the highest, whereas those of (2008), Ülkü and Prodan (2013) uncover that
MACD-3 strategy is significantly unprofitable. moving average convergence divergence (MACD)
However, the predictability of MACD-2 is trading rules have predictive ability in many
significantly bigger than that of MACD-4 strategy countries. There are several MACD trading
(see Table 5). This result is in slight contrast with strategies applied to execute the trading signals.
the results in the base sample where the This study analyzes the performance of a MACD
predictability of MACD-4 strategy isbetter than that trading strategy (MACD-4), which is applied
of MACD-2 strategy (see Table 4). The reason is popularly by practitioners but has not tested by prior
that there is a contrast between HOSE results and academicians. We compare the performance of each
SSE results in the new fresh sample. Whereasall of the strategies on a group of markets to confirm
results from SSE market fully confirm those from the best one.
base sample market, the results from HOSE do not.
Indeed, figures in Table 5 reveal that the highest The results reveal that the MACD-4 strategy is
mean daily buy minus sell returns and mean daily significantly profitable in RTS. However, the
excess returns are produced by MACD-2 strategy profitable returns of the strategy are not significant
with 0.00203 percent and 0.0152 percent, in Nikkei 225 and Dax. We realize that the highest
respectively. Following those are those produced by profitability is earned by the MACD-1 strategy. The
MACD-4 strategy with 0.00177 percent and second is the MACD-4 strategy and the least
0.00126 percent, respectively. The least returns are profitability is earned bythe MACD-2 strategy, on
created by MACD-1 strategy with 0.00171 percent average. Furthermore, the numbers of trades by the
and 0.00121 percent, respectively, in HOSE market. MACD-1 strategy is the lowest among the four
strategies, and thus it implies that this strategy is the
Conclusions most effective one if it is used in a high trading cost
Some studies published recently Dejan Eric (2009), environment. Especially, we find evidence to prove
R. Rosillo (2013), Terence Tai-Leung Chong that MACD-3 strategy is useless in the samples.
References
1. Bessembinder, H. and K. Chan (1995). The profitability of technical trading rules in the Asian stock markets,
Pacific-Basin Finance Journal, 3, pp. 257-284.
2. Dejan Eric, G.A., Srdjan Redzepagic (2009). Application on MACD and RSI indicators as functions of investment
strategy optimization on the financial market, Zbornik Radova Ekonomskog Fakulteta u Rijeci, 27, pp. 171-196.
3. Gunasekarage, A. and D.M., Power (2001). The profitability of moving average trading rules in South Asian stock
markets, Emerging Markets Review, 2, pp. 17-33.
4. Ming-Ming, L. and L. Siok-Hwa (2006). The profitability of the simple moving averages and trading range
breakout in the Asian stock markets, Journal of Asian Economics, 17, pp. 144-170.
5. Nguyen Hoang Hung, Y.Z. (2013). Profitability of Applying Simple Moving Average Trading Rules for the
Vietnamese Stock Market, Journal of Business & Management, 2, pp. 22-31.
6. R., Rosillo, D.D.L.F., J.A.L. Brugos (2013). Technical analisis and the Spanish stock exchange: testing the RSI,
MACD, Momentum and stochastic rules using Spanish market companies, Applied Economics, 42, pp. 1541-1550.
7. Sullivan, R., Timmermann, A. and White, H. (1999). Data-Snooping, technical trading, rule performance and
bootstrap, Journal of Finance, 54, pp. 1647-1691.
8. Terence Tai-Leung Chong, S.H.-S.C., Elfreda Nga-Yee Wong (2010). A comparision of Stock Market Efficiency
of the BRIC countries, Technology and Investment, 1, pp. 235-238.
9. Terence Tai-Leung Chong, W.-K.N. (2008). Technical analysis and London stock exchange: testing the MACD
and RSI rules using FT30, Applied Economics Letters, 15, pp. 1111-1114.
10. Ülkü, N. and E. Prodan (2013). Drivers of technical trend-following rules' profitability in world stock markets,
International Review of Financial Analysis, 30, pp. 214-229.
11. William Brock, J.L., Blake LeBaron (1992). Simple Technical Trading Rule and The Stochastics Properties of
Stock Returns, Journal of Finance, 47, pp. 1731-1764.
12. Yu, H., G.V., Nartea, C., Gan and L.J., Yao (2013). Predictive ability and profitability of simple technical trading
rules: recent evidence from Southeast Asian stock markets, International Review of Economics & Finance, 25,
pp. 356-371.

367
Investment Management and Financial Innovations, Volume 13, Issue 2, 2016

Appendix
Table 1. Various MACD trading strategies conducted by researchers on stock markets
MACD trading
Authors Studies titles Various MACD trading strategies
strategies
Terence Tai- Technical analysis and
MACD crosses Leung Chong, London stock exchange: Buy signal is generated when the MACD crosses above the zero line.
MACD-1
the zero line. Wing-Kam Ng testing the MACD and RSI Sell signal is generated when the MACD crosses below the zero line.
(2008) rules using FT30.
Dejan Eric, Application on MACD and
Goran Andjelic, RSI indicators as functions
Srdjan of investment strategy
Redzepagic optimization on the financial
(2009) market.
Terence Tai-
MACD line
Leung Chong, Buy signal is generated when the MACD crosses above the signal line.
MACD-2 crosses signal A comparison of stock
Sam Ho-Sum Sell signal is generated when MACD crosses below the signal line.
line. market efficiency of the
Cheng, Elfreda
BRIC countries.
Nga-Yee Wong
(2010)
Numan Ulkü, Drivers of technical trend-
Eugeniu Prodan following rules' profitability in
(2013) the world stock markets.
Technical analysis and the
MACD line
R. Rosillo, Spanish stock exchange: Buy signal is generated when both the MACD and signal line cross
crosses zero line
D. de la Fuente, testing the RSI, MACD, below the zero line, besides signal line is greater than the MACD. Sell
MACD-3 and MACD line
J.A. L. Brugos momentum and stochastic signal is generated when both the MACD and signal line cross above
crosses signal
(2013) rules using Spanish market the zero line, besides signal line is less than the MACD.
line.
companies.

Table 2. Statistics summary for daily returns

Markets Daily mean St. dev. Skewness Kutoris AC(1) AC(1) AC(1) AC(1) AC(1)
DAX 0.00062 0.016 0.028 4.2 -0.14 -0.15 -0.42 ** 0.43* -0.53*
NIKKEI 225 -0.000093 0.0158 -0.4 6.7 -0.031** -0.019 -0.003 -0.031 -0.025
RTS 0.00077 0.022 -0.47 10.12 0.098* 0.015* -0.043* 0.013* 0.004*
SSE 0.000094 0.016 -0.109 4.34 0.001 -0.015 0.038 0.046 * 0.003*
HOSE 0.0005 0.017 -0.36 3.4 0.29 * 0.042 * 0.023 * 0.101 * 0.117*

Notes: AC(i) refers to the i-th order autocorrelation coefficient; * implies significance at 5% level; ** implies significance at 10%
level.
Table 3. Results of MACD-4 trading strategy
1 2 3 6 7 8 9 10 11
Types of Unconditional Number of Number of Buy-sell Excess
Markets Buy returns Sell returns
MACD rule daily returns buys sells returns returns
0.00060* -0.00019 0.00080*
RTS 0.00077 92 55 0.00002
(3.3) (-0.87) (2.78)
0.00003 -0.00015 0.00018
MACD-4 DAX 0.00006 99 56 0.00012
(0.29) (-0.78) (0.82)
0.00006 -0.00009 0.00015
NIKKEI 225 -0.00009 78 67 0.00025
(-0.16) (-0.15) (0.0002)
Average 0.00028 90 59 0.00023 -0.00014 0.00038 0.00013

Notes: * implies significance at 5% level; ** implies significance at 10% level.

Table 4. Results of the four MACD trading strategies in the base sample
1 2 3 6 7 8 9 10
Types of Unconditional daily Number of Buy-sell
Markets Buy returns Sell returns Excess returns
MACD rules returns trades returns
0.00085* -0.00014 0.00100**
RTS 0.00077 89 0.00022
(3.353) (-0.47) (2.51)
0.00025 -0.00020 0.00045
MACD-1 DAX 0.00006 89 0.00039
(1.47) (-0.83) (1.52)
0.00010 -0.00018 0.00028
NIKKEI 225 -0.00009 90 0.00037
(0.027) (-0.47) (0.36)
Average 0.00028 89 0.00040 -0.00018 0.00058 0.00033
0.00072* -0.00002 0.00074*
MACD-2 RTS 0.00077 227 -0.00004
(2.62) (-0.77) (1.86)

368
 Investment Management and Financial Innovations, Volume 13, Issue 2, 2016

Table 4 (cont.). Results of the four MACD trading strategies in the base sample
1 2 3 6 7 8 9 10
Types of Unconditional daily Number of Buy-sell
Markets Buy returns Sell returns Excess returns
MACD rules returns trades returns
0.00014 -0.00009 0.00023
DAX 0.00006 243 0.00017
(0.75) (-0.39) (0.78)
0.00005 -0.00013 0.00018
NIKKEI 225 -0.00009 218 0.00027
(-0.13) (-0.29) (0.1)
Average 0.00028 229 0.00030 -0.00008 0.00038 0.00014
-0.00019 0.00060* -0.00080*
RTS 0.00077 147 -0.00157
(-0.87) (3.3) (-2.8)
-0.00015 -0.00018
MACD-3 DAX 0.00006 155 0.00003 (0.29) -0.00024
(-0.78) (-0.82)
-0.00009 0.00006 -0.00015
NIKKEI 225 -0.00009 145 -0.00006
(-0.70) (-0.77) (-0.99)
Average 0.00028 149 -0.00014 0.00023 -0.00038 -0.00062
0.00060* -0.00019 0.00080*
RTS 0.00077 147 0.00002
(3.3) (-0.87) (2.78)
0.00003 -0.00015 0.00018
MACD-4 DAX 0.00006 155 0.00012
(0.29) (-0.78) (0.82)
0.00006 -0.00009 0.00015
NIKKEI 225 -0.00009 145 0.00025
(-0.16) (-0.15) (0.0002)
Average 0.00028 149 0.00023 -0.00014 0.00038 0.00013

Notes: * implies significance at 5% level; ** implies significance at 10% level.

Table 5. Results of the four MACD trading strategies in the new fresh sample
1 2 3 6 7 8 9 10
Types of Unconditional daily Number of Buy-sell
Markets Buy returns Sell returns Excess returns
MACD rules returns trades returns
0.00048* -0.00043* 0.00092*
SSE 0.00003 76
(2.56) (-1.95) (3.14) 0.00089
MACD-1
0.00106* -0.00066* 0.00171*
VN-index 0.00051 80
(5.2) (-3.25) (6.01) 0.00121
Average 78 0.00077 -0.00055 0.00132 0.00105
0.00027** -0.00022 0.00049**
SSE 0.00003 228
(1.38) (-1) (1.67) 0.00046
MACD-2
0.00122* -0.00081* 0.00203*
VN-index 0.00051 205
(6.4) (-3.8) (7.12) 0.00152
Average 217 0.00074 -0.00052 0.00126 0.00099
-0.00021 0.00040* -0.00061*
SSE 0.00003 145
(-1.2) (2.97) (-2.77) -0.00064
MACD-3
-0.00067* 0.00110* -0.00177*
VN-index 0.00051 132
(-4.12) (7.1) (-7.9) -0.00227
Average 139 -0.00044 0.00075 -0.00119 -0.00146
0.00040* -0.00021 0.00061*
SSE 0.00003 145 0.00058
(2.97) (-1.2) (2.77)
MACD-4
0.00110* -0.00067* 0.00177*
VN-index 0.00051 132 0.00126
(7.1) (-4.1) (7.9)
Average 139 0.00075 -0.00044 0.00119 0.00092

Notes: * implies significance at 5% level; ** implies significance at 10% level.

369