Modeling Exchange Rate Volatility of ASEAN Member
Countries
PIYASIRI KONGWIRIYAPISAL
School of Economics, Sukhothai Thammathirat Open University
9/9 Moo 9, Bang Phut Subdistrict, Pak Kret District, Nonthaburi 11120
THAILAND
Abstract: - This study investigates the volatility of exchange rates in nine selected ASEAN member
countries, using five forms of the GARCH model. Daily data was sourced from the Bank of Thailand
website database, as Baht per foreign currency, over the period from October 2, 2018 to October 7,
2022. This data included Malaysia Ringgit, Singapore Dollar, Brunei Darussalam Dollar, Philippines
Peso, Indonesia Rupiah, Myanmar Kyat, Cambodia Riel, Laos Kip, and Vietnam Dong. According to
the findings of this study, only eight of the exchange rates were suitable for analysis. In addition, the
GARCH ( 1,1) , TGARCH ( 1,1) , and PGARCH ( 1,1) models were determined to be the most
applicable, and leverage effects were observed in certain exchange rates. To mitigate the risk
associated with trade and investment activities, investors should closely monitor news that is likely to
affect the value of exchange rates. In order to design actions that promote exchange rate stability,
government agents, on the other hand, must ensure they are current on such news.
Key-Words: - Exchange rates, Volatility modeling, ARMA, ARCH- type models, ASEAN member
countries
Received: September 9, 2022. Revised: May 24, 2023. Accepted: June 25, 2023. Published: July 12, 2023.
1.
Introduction
Financial liberalization of the late 20th century
caused dramatic fluctuations in international
currencies and financial assets. This high degree of
volatility has made it difficult to predict the future
and has also raised the associated risks for
investors. To reduce the risk of investing, it is
necessary for investors to accurately evaluate the
price of the assets and value of currencies [1].
Volatility in exchange rates is a factor that has
far- reaching macroeconomic implications,
including impacts on international trade,
investment [ 2] , inflation, and economic growth,
which are all indicators of economic stability and
prosperity. As a result, governments strive to
implement policies to control this volatility in order
to advance economic objectives.
The 1997 Asian financial crisis raised serious
concerns about the economic interdependence,
investment flows, and exchange rate instability of
the Association of Southeast Asian Nations
( ASEAN) member nations. To address these
concerns, some ASEAN nations shifted from
controlled floating regimes with no fixed path for
currency rates to stable arrangements, while others
moved to floating regimes. However, it was reveal
that despite the different exchange rate management
regimes in this region, the real exchange rates of the
ASEAN currencies follow similar cycles and trends
over the long term, indicating the interconnection
between countries [3] and it was recognize that the
Thai baht served as the primary conduit through
which regional currency fluctuations were
transmitted [4].
Previous works have highlighted the potential
negative impacts of exchange rate volatility on
countries. For example, [ 5] found that countries
with higher exchange rate volatility are more likely
to experience a trade deficit, while [ 6] found that
countries with higher exchange rate volatility were
more likely to experience a decrease in exports.
Similarly, [ 7] found that countries with higher
exchange rate volatility were more likely to
INTERNATIONAL JOURNAL OF APPLIED MATHEMATICS,
COMPUTATIONAL SCIENCE AND SYSTEMS ENGINEERING
DOI: 10.37394/232026.2023.5.8
Piyasiri Kongwiriyapisal
E-ISSN: 2766-9823
84
Volume 5, 2023
experience a decrease in export performance. In
contrast, [ 8] could not confirm the impact of
exchange rate volatility on import. For the
investment, research conducted by [9] indicates that
the effect of exchange rate volatility on foreign
direct investment ( FDI) may differ depending on
the context.
ASEAN countries, in particular, may
experience substantial effects of exchange rate
volatility on trade balances [ 10] , foreign direct
investment [ 11] , economic growth rate [ 12] ,
imports and exports [ 13] , as well as domestic
consumer and producer prices [14].
For the benefit of policy design, academics
frequently employ ARCH- type models, such as
ARCH, GARCH, TGARCH, EGARCH, PGARCH,
SGARCH, and TCHARCH, to examine the volatility
of the exchange rate. This research therefore aims to
apply the ARCH family model to investigate the
exchange rate volatility of ASEAN member countries.
The results of this research are set to aid investors and
governments in designing strategies and policies to
prevent risk and foster economic and business growth
and stability. The work is structured as follows: Section
2 presents the academic works that support the research
framework. Section 3 then outlines the research
methodology. Section 4 provides the results and
discussion.
2.
Literature Review
Exchange rate volatility is a phenomenon that has
intrigued researchers for decades. While the exact
cause of such extreme fluctuations has yet to be
determined, multiple explanations have been
suggested to explain this phenomenon, e. g. ,
historical information and leverage effect [15].
In the case of the leverage effect, a negative
correlation between return and volatility has been
observed in the stock market [ 16] . This effect is
expected to persist in the foreign exchange market
as well, as it has been established that news,
especially bad news, can have a substantial effect
on the conditional volatility of exchange rates when
they are closely intertwined [ 17] . This is because
the volatility of one currency can spill over to other
currencies, which can lead to significant losses for
investors.
Utilizing ARCH- type models has been a
prominent method for analyzing exchange rate
volatility from various perspectives and identifying
the model that best describes the characteristics of
exchange rate volatility. In order to gain a deeper
understanding of the ARCH-type models in current
research, the following will present an overview of
prior empirical studies employing these models.
In [ 18] , the scholars investigated the nature of
exchange rate volatility in the exchange rates of the
Vietnamese dong (VND). Using ARMA-GARCH
models to capture the mean and volatility process of
USD- VND, GBP- VND, JPY- VND, and CAD-
VND exchange rate returns, the author found that
these models were well-adequate. In [19], they also
investigated exchange rate volatility, finding that
the GARCH model was the best model to explain
the volatility of the return on the exchange of
Afghanistan's foreign exchange rate with the US
dollar. In [ 20] , the scholars conducted a similar
investigation into Southeast Asian countries,
finding that the PARCH model was appropriate for
Malaysian Ringgit (MYR), Vietnam Dong (VND),
and Singapore Dollar ( SGD) , while the GARCH
model was appropriate for THB and PHP, and the
TARCH model was appropriate for Indonesia
Rupiah (IDR). These findings indicate that different
exchange rate volatility models may be better suited
for different currencies.
The following academics' works contain
evidence for the leverage effect: in [ 21] , they
conducted a study on the Pakistan- US dollar
exchange rate volatility and found a negative
significance of the EGARCH's leverage value. In
[ 22] , the scholar examined the volatility of the
RMB exchange rate return for both onshore and
offshore markets and revealed the presence of
leverage effects in both. In [23], they found that the
best-fitting model for China, India, Spain, UK, and
the USA is GJR- GARCH, followed by GARCH,
TGARCH, and EGARCH. Also, in [ 24] , the
scholars modeled the volatility of Somali shilling
against US dollar and found that the TCHARCH
and EGARCH models were more suitable, with
evidence of a leverage effect - positive news having
a greater effect on volatility than bad news of the
same magnitude.
Numerous studies have been conducted in an
INTERNATIONAL JOURNAL OF APPLIED MATHEMATICS,
COMPUTATIONAL SCIENCE AND SYSTEMS ENGINEERING
DOI: 10.37394/232026.2023.5.8
Piyasiri Kongwiriyapisal
E-ISSN: 2766-9823
85
Volume 5, 2023
effort to better comprehend the effects of historical
data and market news on the foreign exchange
market. These studies demonstrate that various
exchange rates are fitted with models that reflect the
diverse effects of historical data and market news
on the exchange rate. The most important takeaway
from these previous studies is that different
exchange rates are fitted with various models that
reflect the diverse effects of historical data and
market news on the foreign exchange market.
3.
Methodology
The ARMA model, also known as the Box-Jenkins
methodology, encompasses autoregressive and
moving average methods. This model estimates the
dependent variable by using its own lags and past
errors. The proper number of lags and past errors,
determined by ACF and PACF, that are included in
the model is often evaluated by the AIC criteria,
which indicates the proper model by the smallest
AIC value. The details of this model are as
follows.
AR Model
An autoregressive model is based on the idea that
the current value of the dependent variable can be
explained by its past values. This model can be
presented by:
1
p
t i t i t
i
YY
, (1)
where
is a constant.
i
,
1,2,...,ip
, is the
parameter of the model AR,
p
is the order of the
model, and
t
represents the error that cannot be
explained by the model.
MA Model
A moving model is based on the idea that the
current value of the dependent variable can be
explained by past errors. This model can be written
by:
1
q
t j t j t
j
Y
, (2)
where
is a constant.
j
,
1,2,...,jq
, is the
parameter of the model and
q
is the order of the
model.
ARMA Model
The ARMA(p, q) model is the combination of the
above two models. If
t
Y
is stationary, this model
can be represented by:
11
pq
t i t i j t j t
ij
YY
, (3)
where
is a constant.
ARCH-type Models
The autoregressive conditional heteroscedasticity
(ARCH) family model can be used to justify the
volatility of price and return in the financial market.
This model enables the analysts to trace the patterns
of market fluctuation. To understand how the model
within the ARCH family was formed for this study,
this section will provide a brief overview of five
models used in this research. The details of each
model are as follows.
ARCH model.
The Autoregressive Conditionally Heteroscedastic
Model, ARCH ( q) , is used to describe the variance
of the current error term. This model is commonly
applied in modeling financial time series that
exhibit time- varying volatility and volatility
clustering, and it can be stated as follows:
2 2 2
0 1 1 2 2 ...
t q t q
tt
h
, (4)
2
0
1
q
t i t i
i
h
, (5)
where
t
h
is the conditional variances.
t
denotes the
error term.
q
is the number of lags.
00
and
0, 1,...,
iiq
. This implies that the
conditional variance depends on previously squared
residuals and needs to be non-negative. If
0
i
,
t
h
will equals to constant and thus conditional
variance is homoscedastic.
INTERNATIONAL JOURNAL OF APPLIED MATHEMATICS,
COMPUTATIONAL SCIENCE AND SYSTEMS ENGINEERING
DOI: 10.37394/232026.2023.5.8
Piyasiri Kongwiriyapisal
E-ISSN: 2766-9823
86
Volume 5, 2023
GARCH Model
Generalized autoregressive conditional
heteroscedastic models, GARCH ( p, q) , allow for
both a long memory and a more flexible lag
structure. While ARCH models only concern that
the conditional variance is linearly associated with
the past variances, GARCH (p, q) models added the
previous conditional variances into the model. The
p and q in the model denote the GARCH element
and the ARCH element, respectively. The
specification of the GARCH ( p, q) process is as
follows:
2
0
11
qp
t i t i i t j
ij
hh
, (6)
where
p
is the number of lags.
0
j
.
1t
is an
ARCH term that represents a previous shock.
1t
h
is a GARCH term which represents the past
forecasted conditional variance.
Threshold GARCH Model
The Threshold Autoregressive Conditionally
Heteroscedastic Model, TARCH, uses a piecewise
equation for the conditional standard deviation to
permit asymmetry in the conditional variance. The
TARCH model is mathematically defined as:
22
0
1 1 1
qp
r
t i t i j t j
k t k t k
i k j
Ihh
, (7)
where
1
t
I
if
0
t
and 0 otherwise. In this
model, good news,
0
ti
, and bad news.
0
ti
, have differential effects on the conditional
variance; good news has an impact of
i
, while
bad news has an impact of
ii
. If
0
i
, bad
news increases volatility, and we say that there is a
leverage effect for the i- th order. If
0
i
, the
news impact is asymmetric.
Exponential GARCH Model
The Exponential Generalized Autoregressive
Conditionally Heteroscedastic Model, EGARCH,
controls asymmetry in financial data. Even if the
estimated coefficients are negative, the logarithmic
characteristics of the EGARCH model guarantee
that the conditional variance is positive. The
expression of the conditional variance for an
EGARCH model is stated as follows:
22
0
1 1 1
log log
qp
r
t i t k
t i j t j
k
i k j
ti tk
hh
hh
,
(8)
The left-hand side is the log of the conditional
variance. This implies that the leverage effect is
exponential, rather than quadratic, and that
forecasts of the conditional variance are guaranteed
to be nonnegative. The presence of leverage effects
can be tested by the hypothesis that
0
k
. The
impact is asymmetric if
0
k
.
Power GARCH model
PGARCH is modeled by standard deviation rather
than variance. The PARCH model may be specified
as follows:
0
11
t i i t i
qp
ij
ij
tj
t
hh
, (9)
where
0
,
1
i
for
1, 2,...,ir
.
0
i
,
for all
ir
and
rp
. The optional
i
parameters are added to capture asymmetry of up to
order
r
. If
2
and
0
i
, the PARCH model is
simply a GARCH model. The asymmetric effects
are present if
0
k
.
To evaluate the error of the results from the
model estimation, root-mean-square error (RMSE)
and mean absolute error ( MAE) will be used, and
they can be written as follows:
2
1
1
RMS ˆ
E
N
ii
i
Ne e
, (10)
1
1ˆ
N
ii
i
ME eAe
N
, (11)
where N is the sample number,
e
is the actual
exchange rate return, and
ˆ
e
is the forecast
exchange rate return.
INTERNATIONAL JOURNAL OF APPLIED MATHEMATICS,
COMPUTATIONAL SCIENCE AND SYSTEMS ENGINEERING
DOI: 10.37394/232026.2023.5.8
Piyasiri Kongwiriyapisal
E-ISSN: 2766-9823
87
Volume 5, 2023
The exchange rate data used for estimating
the model was downloaded from the Bank of
Thailand website database as Baht/ Foreign
currency. These data are daily basic which include
Baht/ Malaysia Ringgit ( MYR) , Baht/ Singapore
Dollar ( SGD) , Baht/ Brunei Darussalam Dollar
(BND) , Baht/ Philippines Peso ( PHP) ,
Baht/ Indonesia Rupiah ( 1000 Rupiah) ( IDR) ,
Baht/Myanmar Kyat (MMK), Baht/Cambodia Riel
(100 Riel)(KHR), Baht/Laos Kip (100 Kip)(LAK),
and Baht/ Vietnam Dong ( VND100)(VND) . The
data covers the time between October 2, 2018 and
October 7, 2022, including 1,015 observations.
4.
Result and Discussion
The Unit Root Tests of the exchange rate are
reported in Table 1 which indicates the stationary of
the exchange rates.
Table 1: Augmented Dickey-Fuller Unit Root Tests
Var.
t-Stat.
Prob.
LDBND
-27.246
0.000
LDIDR
-28.528
0.000
LDKHR
-31.712
0.000
LDLAK
-29.527
0.000
LDMMK
-30.879
0.000
LDMYR
-28.273
0.000
LDPHP
-26.145
0.000
LDSGD
-26.798
0.000
LDVND
-29.534
0.000
In order to examine the ARCH effect, the
ARMA model for exchange rate returns was
identified, and the result of analysis reveals that the
ARCH effect does not persist in LDMMK, ARMA
( 2,2) , ARCH( 1) ( F- stat. = 0. 342428, P-
value=0.558). Therefore only 8 exchange rates, i.e.,
LDBND, LDIDR, LDKHR, LDLAK, LDMYR,
LDPHP, LDSGD, and LDVND, will be further
investigated. In this study the appropriate models
will be selected based on AIC criteria such that the
lowest value is the best. From the analysis it show
the appropriated models for each exchange rate as
follows: LDBND ARMA ( 1,0) TGARCH ( 1,2)
( AIC= -9. 0760) ; LDIDR ARMA ( 2,3) PGARCH
( 1,1) ( AIC= - 8. 0025) ; LDKHR ARMA ( 3,3)
GARCH ( 1,1) ( AIC= 8. 2041) ; LDLAK ARMA
( 3,1) PGARCH ( 1,1) ( AIC= - 8. 1935) ; LDMYR
ARMA ( 1,0) PGARCH ( 1,1) ( AIC= - 8. 7470) ;
LDPHP ARMA ( 1,0) TGARCH ( 1,2) ( AIC= -
8. 5894) ; LDSGD ARMA ( 1,0) GARCH ( 1,1)
( AIC= 9. 0961) ; and LDVND ARMA( 1,0)
TGARCH ( 1,2) ( AIC= - 8. 5482). The results of
model estimations are shown in Table 2.
From LDBND TGARCH ( 1,2) in Table 2, it
was found that the coefficients of the asymmetric
parameter are significant. This shows that the
leverage effect is present in this exchange rate
and leads to the conclusion that LDBND reacts
more to good news than to bad news.
Table 2: Model estimations
LDBND
TGARCH
(1,2)
LDIDR
PGARCH
(1,1)
LDKHR
GARCH
(1,1)
LDLAK
PGARCH
(1,1)
0
1.1E-07**
5.4E-06
2.6E-07
3.9E-10
0.050***
0.232***
0.038***
0.034***
1
0.100**
0.062
-
-0.197***
2
-0.129***
-
-
-
0.431***
-
0.953***
0.948***
2.046***
0.946***
2.760***
LDMYR
PGARCH
(1,1)
LDPHP
TGARCH
(1,2)
LDSGD
GARCH
(1,1)
LDVND
TGARCH
(1,2)
0
0.000
1.5E-06**
1.0E-07**
6.5E-08**
0.043***
0.097***
0.037***
0.059***
1
-0.583***
0.048
0.074
2
-0.090**
-0.124**
0.947***
0.800**
0.784***
0.947***
0.962***
For LDIDR PGARCH ( 1,1) , it demonstrates
that the asymmetric coefficient is not statistically
significant, establishing the nonexistence of the
leverage effect and implying that GARCH ( 1,1)
is more suitable. According to LDKHR GARCH
INTERNATIONAL JOURNAL OF APPLIED MATHEMATICS,
COMPUTATIONAL SCIENCE AND SYSTEMS ENGINEERING
DOI: 10.37394/232026.2023.5.8
Piyasiri Kongwiriyapisal
E-ISSN: 2766-9823
88
Volume 5, 2023
( 1,1) , all coefficients are significant. Thus, both
past conditional variances and errors influence
the current conditional variance. LDLAK
PGARCH ( 1,1) demonstrates that the
asymmetric coefficient is statistically significant,
proving the existence of the leverage effect and
indicating that this exchange rate reacts more to
positive than negative news.
The remaining model estimation results can be
interpreted in a similar manner. The results led us
to the conclusion that the leverage effect exists in
LDBND, LDLAK, LDMYR, LDPHP, and
LDVND, indicating that these exchange rates are
responsive to news on the foreign exchange market.
The properties of the models, i. e. , serial
correlation, ARCH effect, and normal distribution
of residuals, are investigated based on the following
hypotheses:
0
H
: there is no serial correlation in the
residual;
0
H
: there is no ARCH; and
0
H
: residuals
are normally distributed. The results from this
investigation reveal that all models present no serial
correlation and no ARCH effect. However, the
residuals are not normally distributed.
The results of the volatility estimation in Figure
1 indicate that DLLAK, DLMYR, DLSGD, and
DLVND not only have high volatility but also a
rising trend. Table 3 displays the forecast error.
The results of the analysis indicate that market
shocks and historical information influence
exchange rate volatility. In addition, the
significance of coefficients and leverage effect
parameters can be utilized to divide the exchange
rate into two groups: those that are consistent with
a symmetric model and those that are consistent
with an asymmetric model.
The exchange rates in the symmetric group
include LDKHR, LDSGD, and LDIDR. These
exchange rates fit with GARCH models, which
have been used in a number of recent studies. For
example, Mahroowal & Salari ( 2019) who used a
GARCH model to explain the volatility of the return
on the exchange of Afghanistan's foreign exchange
rate; Nguyen (2018) who used a GARCH model to
explain the volatility of USD- VND, GBP- VND,
JPY-VND, and CAD-VND exchange rate returns;
and SEKMEN & Ravanoğlu ( 2020) who used a
GARCH model to explain the volatility of the
selected exchange rates.
In the case of the asymmetric group, it consists
of LDBND, LDLAK, LDMYR, LDPHP and
LDVND. These exchange rates contain the leverage
effect expressed by the TGARCH and PGARCH
models. Recent studies that used these models to
estimate exchange rate volatility include, e.g., the
work of Ponziani ( 2019) and Rehman & Salamat
( 2021) , who indicated the existence of an
asymmetric effect of good news and bad news on
exchange rate volatility.
LDBND
DIDR
LDKHR
LDLAK
LDMYR
LDPHP
LDSGD
LDVND
Figure 1: exchange rate volatility estimations
Table 3: Errors of exchange rate estimation
Ex. rate
RMSE
MAE
LDBND
0.0026
0.0020
LDIDR
0.0046
0.0035
LDKHR
0.0040
0.0031
LDLAK
0.0045
0.0031
LDMYR
0.0031
0.0024
LDPHP
0.0033
0.0026
LDSGD
0.0026
0.0020
LDVND
0.0036
0.0026
INTERNATIONAL JOURNAL OF APPLIED MATHEMATICS,
COMPUTATIONAL SCIENCE AND SYSTEMS ENGINEERING
DOI: 10.37394/232026.2023.5.8
Piyasiri Kongwiriyapisal
E-ISSN: 2766-9823
89
Volume 5, 2023
Given the preceding information, investors in
LDBND, LDLAK, LDMYR, LDPHP, and LDVND
should maintain current information, search for
news that may affect the volatility of these
exchange rates, and restructure their investments to
avoid losing money due to the volatility risk
associated with these rates. However, investors can
reduce the risk of volatility through prudent
investment planning. This can be accomplished by
selecting appropriate investment currencies. In
addition, investors can employ hedging strategies,
such as the use of options, to reduce their risk of
volatility-related losses (Bhunia & Ganguly, 2020).
This is also significant for central banks, as
these exchange rates are integral to the businesses
and economic activities that rely on them. A
significant change in these exchange rates could
have far- reaching effects on the economy.
Therefore, central banks should have contingency
plans for any potential shocks.
5.
Managerial Implication
Based on the findings, this research suggests that
investors who are interested in investing in foreign
currencies investigated in this research and
government agents who respond to the stability of
macroeconomic indicators, e. g., aggregated price,
trade balance, GDP growth rate, and value of Baht,
be prepared to deal with the potential risk from
exchange rate volatility. They should pay special
attention to LDBND, LDLAK, LDMYR, LDPHP,
and LDVND as their volatility is sensitive to the
news. Therefore, searching for information about
the expected value of the Baht should be a good
strategy for investors to prevent risk from trade and
investment activities and for government agents to
design actions to intervene in the exchange market
to foster exchange rate stability.
6.
Limitations and Future Research
This study has only demonstrated the impact of
historical data and leverage on the volatility of
ASEAN Member Countries' exchange rates.
Additionally, only five ARCH- type models were
utilized in this study. Consequently, future research
may take into account additional alternative time
series models to address volatility and utilize
various forms of ARCH- type models. Inflation,
interest rates, and international reserves may be
considered variables.
7. Conclusion
In this study, the characteristics of the daily
exchange rate returns of the 9 selected ASEAN
member countries are examined. The 8 forms of the
GARCH models, i. e. , ARCH ( 1) , ARCH ( 2) ,
GARCH ( 1,1) , GARCH ( 1,2) , TGARCH ( 1,1) ,
EGARCH ( 1,1) , and PGARCH ( 1,1) are used to
address the volatility of these exchange rates. The
exchange rate data used for estimating the model
was downloaded from the Bank of Thailand website
database as Baht/ Foreign currency. These data are
daily basic which include Baht/ Malaysia Ringgit
( MYR) , Baht/ Singapore Dollar ( SGD) ,
Baht/ Brunei Darussalam Dollar ( BND) ,
Baht/ Philippines Peso ( PHP) , Baht/ Indonesia
Rupiah ( IDR) , Baht/ Myanmar Kyat ( MMK) ,
Baht/ Cambodia Riel( KHR) , Baht/ Laos Kip, and
Baht/ Vietnam Dong ( VND) . The data covers the
time between October 2, 2018 and October 7, 2022.
Before analyzing the volatility, these exchange
rates are manipulated by the log of the first
difference. The appropriate models are selected
based on the AIC criteria. After consideration, it
was discovered that BND matched TGARCH (1,1),
IDR matched PGARCH ( 1,1) , KHR matched
GARCH ( 1,1) , LAK matched PGARCH ( 1,1) ,
MYR matched PGARCH ( 1,1) , PHP matched
TGARCH (1,1), SGD matched GARCH (1,1), and
VND matched TGARCH ( 1,1) . The model
estimation shows that all models present no serial
correlation and no ARCH effect. However, the
residuals are not normally distributed. In addition,
there are leverage effects in BND, LAK, MYR,
PHP, and VND. The result of volatility estimation
shows that LAK, MYR, SGD, and VND not only
have high volatility but also an increasing trend.
Therefore, investors should search for news relating
to these exchange rates to prevent risk from trade
and investment activities. Also, government agents
need to search for such news to design actions to
intervene in the exchange market to foster exchange
rate stability.
INTERNATIONAL JOURNAL OF APPLIED MATHEMATICS,
COMPUTATIONAL SCIENCE AND SYSTEMS ENGINEERING
DOI: 10.37394/232026.2023.5.8
Piyasiri Kongwiriyapisal
E-ISSN: 2766-9823
90
Volume 5, 2023
Acknowledgement
This research is supported by School of Economics,
Sukhothai Thammathirat Open University,
Thailand.
References
[1] L. Kantar, “ ARCH Models and an
Application on Exchange Rate Volatility:
ARCH and GARCH Models,” in Handbook
of Research on Emerging Theories, Models,
and Applications of Financial Econometrics,
Springer, 2021, pp. 287–300.
[2] M. A. Rehman and D. A. Salamat, “Modeling
exchange rate volatility with GARCH
models: A comparison based on a volatility
breaks,” Asian Finance Research Journal,
vol. 3, no. 1, pp. 1–15, 2021.
[3] I. Sangaré, “ Exchange Rate Regimes in the
ASEAN: Would a Currency Union
Outperform the Independent Managed
Floating Regimes?,” Journal of Economic
Integration, vol. 36, no. 1, pp. 72–102, Mar.
2021, doi: 10.11130/jei.2021.36.1.72.
[4] D. T. Hurley and R. A. Santos, “ Exchange
rate volatility and the role of regional
currency linkages: the ASEAN case,” Applied
Economics, vol. 33, no. 15, pp. 1991– 1999,
Dec. 2001, doi:
10.1080/00036840010023779.
[5] D. B. Ewubare and C. D. Merenini, “ The
effect of exchange rate fluctuation on foreign
trade in Nigeria,” International Journal of
Scientific Research and Engineering
Development, vol. 2, no. 1, pp. 68–85, 2019.
[6] T. Mosbei, S. K. Samoei, C. C. Tison, and E.
K. Kipchoge, “Exchange rate volatility and its
effect on intra- East Africa Community
regional trade,” Latin American Journal of
Trade Policy, vol. 4, no. 9, pp. 43–53, 2021.
[7] E. Ekanayake and A. Dissanayake, “Effects
of Real Exchange Rate Volatility on Trade:
Empirical Analysis of the United States
Exports to BRICS,” Journal of Risk and
Financial Management, vol. 15, no. 2, pp. 1–
21, 2022.
[8] A. -R. B. Yussif, S. T. Onifade, A. Ay, M.
Canitez, and F. V. Bekun, “ Modeling the
volatility of exchange rate and international
trade in Ghana: empirical evidence from
GARCH and EGARCH,” JEAS, pp. 1– 17,
2022, doi: 10.1108/JEAS-11-2020-0187.
[9] A. Heroja, “ Exchange Rate Effects on
Foreign Direct Investment Decision,” 2022.
[10] C.-H. Puah, S.-W. Yong, S. A. Mansor, and
E. Lau, “ Exchange Rate and Trade Balance
Nexus in ASEAN- 5,” Labuan Bulletin of
International Business & Finance, vol. 6, pp.
19–37, 2008.
[11] J. Lily, M. Kogid, D. Mulok, L. Thien Sang,
and R. Asid, “Exchange Rate Movement and
Foreign Direct Investment in Asean
Economies,” Economics Research
International, vol. 2014, pp. 1–10, Mar. 2014,
doi: 10.1155/2014/320949.
[12] P. T. H. An, N. T. Binh, and H. L. N. Cam,
“ The Impact of Exchange Rate on Economic
Growth - Case Studies of Countries in the
Asean Region,” vol. 9, no. 7, pp. 965– 970,
2018.
[13] K. P. Upadhyaya, D. Dhakal, and F. G.
Mixon, “ Exchange Rate Volatility and
Exports: Some New Estimates From the
Asean-5,” The Journal of Developing Areas,
vol. 54, no. 1, pp. 65– 73, 2020, doi:
10.1353/jda.2020.0004.
[14] V. T. Anh, L. T. T. Quan, N. V. Phuc, H. M.
Chi, and V. H. Duc, “ Exchange Rate Pass-
Through in ASEAN Countries: An
Application of the SVAR Model,” Emerging
Markets Finance and Trade, vol. 57, no. 1,
pp. 21– 34, Jan. 2021, doi:
10.1080/1540496X.2018.1474737.
[15] S. Z. S. Abdalla, “Modelling Exchange Rate
Volatility using GARCH Models: Empirical
Evidence from Arab Countries,” IJEF, vol. 4,
no. 3, pp. 216– 229, 2012, doi:
10.5539/ijef.v4n3p216.
[16] C. Maya and K. Gómez, “ What exactly is’
Bad News’ in foreign exchange markets?:
Evidence from Latin American markets,”
Cuadernos de economía, vol. 45, no. 132, pp.
161–183, 2008.
[17] A. K. Panda, S. Nanda, V. K. Singh, and S.
Kumar, “ Evidence of leverage effects and
volatility spillover among exchange rates of
selected emerging and growth leading
economies,” Journal of Financial Economic
Policy, vol. 11, no. 2, pp. 174–192, 2018.
[18] T. K. D. Nguyen, “Modelling exchange rate
volatility using GARCH model: An empirical
analysis for Vietnam,” presented at the
International Econometric Conference of
Vietnam, 2018, pp. 941–952.
[19] M. Mahroowal and H. Salari, “ Modeling
exchange rate volatility, using Univariate
Generalized Autoregressive conditionally
Hetroscedastic type models: evidence from
INTERNATIONAL JOURNAL OF APPLIED MATHEMATICS,
COMPUTATIONAL SCIENCE AND SYSTEMS ENGINEERING
DOI: 10.37394/232026.2023.5.8
Piyasiri Kongwiriyapisal
E-ISSN: 2766-9823
91
Volume 5, 2023
Afghanistan,” American Journal of
Economics and Business Management, vol. 2,
no. 3, pp. 69–82, 2019.
[20] R. M. Ponziani, “Foreign exchange volatility
modeling of Southeast Asian major
economies,” JEBAV, vol. 22, no. 2, pp. 283–
297, 2019.
[21] M. Mohsin, S. Naseem, D. Muneer, and S.
Salamat, “ The volatility of exchange rate
using GARCH type models with normal
distribution: evidence from Pakistan,” Pacific
Business Review Internationsl, vol. 11, no. 2,
pp. 124–129, 2019.
[22] A. Atabani Adi, “ Modeling exchange rate
return volatility of RMB/USD using GARCH
family models,” Journal of Chinese
Economic and Business Studies, vol. 17, no.
2, pp. 169–187, 2019.
[23] A. Abdulhakeem, S. O. Adams, and R. O.
Akano, “ Modelling Nigeria Naria Exchange
Rate against Some Selected Country’ s
Currencies Volatility: Application of
GARCH Model,” Asian Journal of
Probability and Statistics, vol. 5, no. 1, pp. 1–
13, 2019.
[24] A. O. Ali, “ Modeling Exchange Rate
Volatility of Somali Shilling Against Us
Dollar by Utilizing GARCH Models,” IJEFI,
vol. 11, no. 2, pp. 35– 39, Mar. 2021, doi:
10.32479/ijefi.9788.
Contribution of Individual Authors to the
Creation of a Scientific Article (Ghostwriting
Policy)
The author contributed in the present research, at all
stages from the formulation of the problem to the
final findings and solution.
Sources of Funding for Research Presented in a
Scientific Article or Scientific Article Itself
Conflict of Interest
The author has no conflict of interest to declare that
is relevant to the content of this article.
Creative Commons Attribution License 4.0
(Attribution 4.0 International, CC BY 4.0)
This article is published under the terms of the
Creative Commons Attribution License 4.0
https://creativecommons.org/licenses/by/4.0/deed.en
_US
This research is supported by School of Economics,
Sukhothai Thammathirat Open University,
Thailand.
INTERNATIONAL JOURNAL OF APPLIED MATHEMATICS,
COMPUTATIONAL SCIENCE AND SYSTEMS ENGINEERING
DOI: 10.37394/232026.2023.5.8
Piyasiri Kongwiriyapisal
E-ISSN: 2766-9823
92
Volume 5, 2023
