Application of GARCH Model in Forecasting IDR/USD Exchange
Rate
Khairina Natsir
Economic Faculty, Tarumanagara University, Jakarta, Indonesia
Keywords: Modeling, Exchange Rate, Garch, Forecasting.
Abstract: Modeling and Forecasting the IDR to USD exchange rate is crucial in business as it provides information on
the model of exchange rate fluctuation and taking the right financial decisions. Therefore, financial
managers in a multinational company are required to be able to understand exchange rate forecasting in
order to make financial decisions to optimize the value of the company. The purpose of this research is to do
the modeling and forecasting of IDR exchange rate against USD using GARCH model. The GARCH model
is a suitable model used for financial analysis because assuming the existence of heteroscedasticity not a
problem but can be used to predict future price volatility. GARCH models pay attention to the variance and
errors in doing the forecasting. The results showed that the GARCH model (1,1) was the best model in
representing exchange rate movements during the study period. The result of forecasting of IDR to USD
exchange rate for 5 days after the research period are 14065, 04072, 14078, 14084 and 14090.
1 INTRODUCTION
Globalization has brought about openness in many
ways, including in terms of trade and economics.
Foreign exchange activities or shortened as forex is
often done by all business actors in the world, such
as import export activities, market needs and bank
institutions. Information on exchange rates helps
business people in making investment decisions and
trading their money in order to earn a profit.
Exchange rate forecasting, especially between
IDR to USD is one of the most important aspects in
Indonesia. The exchange rate of IDR/USD is one of
the foundations in the current national economic
activity. The exchange rate is the ratio between the
currency of a country and the currency of another
country. The exchange rate is also one of the most
important macroeconomic variables, because strong
currency exchange rates can maintain economic
stability in an area or country. The economic crisis
that struck Indonesia was preceded by the
emergence of the IDR exchange rate crisis which
was a consequence of an increasingly globally
integrated financial system. This can trigger issues
in financial and banking transaction activities.
Forecasting can minimize the risks that may occur
due to demand uncertainty and others (Natsir and
Mimi, 2017). However, the IDR against USD
exchange rate modeling has not been studied
thoroughly. Through this modeling will provide a
strong signal in the determination of policy and
planning everything related to financial transactions
involving the exchange rate of IDR against USD.
The exchange rate movement of IDR/USD
always fluctuates over time. The high volatility of
the exchange rate makes it difficult to model with
classic OLS, because according to Gauss Markov
theorem, one of the requirements in OLS model is
the variance and error must be constant
(homoscedasticity). This is as such so that the
estimator obtained is BLUE (Hueter and No, 2016).
In this era of globalization, especially in a
floating exchange rate policy, exchange rate
movements will be highly volatile or have high
volatility due to the large number of local or global
factors that affect it. High volatility has the potential
to cause heteroscedastic variance and error.
Therefore, the GARCH model would be more
appropriately used to analyze the exchange rate
because this model does not regard
heteroscedasticity as a constraint, but instead uses
that condition to build the model.
Several studies on exchange rate modeling using
ARCH and GARCH models have attracted the
attention of previous researchers. The study of