scholarly journals Long memory in high frequency foreign exchange rates: Hurst exponents dependence on data aggregation

2010 ◽  
Vol 51 ◽  
Author(s):  
Milda Pranckevičiūtė
Keyword(s):  
Exchange Rates ◽  
Foreign Exchange ◽  
Data Aggregation ◽  
Long Memory ◽  
High Frequency ◽  
Foreign Exchange Rates ◽  
Maximum Price ◽  
The Us ◽  
Japanese Yen ◽  
Hurst Exponents

This paper presents the study on long memory in absolute daily returns of the US dollar versus euro, the British pound and the Japanese yen aggregated foreign exchange rates. Pointwise, maximum price, minimum price and average price aggregation rules for high frequency foreign exchange rates are introduced. The classical R/S statistic is used to analyze Hurst exponents dependence on the choice of data aggregation function.

scholarly journals Modelling foreign exchange rates: a comparison between markov-switching and markov-switching GARCH

2020 ◽  
Vol 20 (2) ◽  
pp. 917
Author(s):  
Mohd Azizi Amin Nunian ◽  
Siti Meriam Zahari ◽  
S.Sarifah Radiah Shariff
Keyword(s):  
Exchange Rates ◽  
Foreign Exchange ◽  
Markov Switching ◽  
Foreign Exchange Rates ◽  
Markov Switching Model ◽  
Switching Model ◽  
The Us ◽  
Log Likelihood ◽  
High Volatility ◽  
Japanese Yen

Foreign exchange rate is important as it determines a country's economic condition. It is used to carry out transfers of purchasing power between two or more countries. Volatility in exchange rates may result in difficulty in decision making especially, in financial sectors as high volatility could increase the risk in exchange rates. Thus, Markov switching model is employed in this study as it is believed to be efficient in handling not only volatilility but also nonlinearity characteristics in exchange rates. The aims of this study are to model the foreign exchange rates using two models; Markov Switching (M-S) models and Markov Switching Generalized Autoregressive Conditional Heteroscedasticity (M-S GARCH) and to compare these two models based on log-likelihood, AIC and BIC criteria. This study used the quarterly data of foreign exchange rates for Singapore Dollar (SGD), Korean Won (KRW), China Yuan Renminbi (CNY), Japanese Yen (JPY) and the US Dollar (USD) against Malaysia Ringgit (MYR) which were collected from Quarter 4, 2006 to Quarter 1, 2018. The findings indicate that Markov Switching is the best model since it has the highest log-likelihood value, and the lowest AIC and BIC values. The results show that JPY and SGD have highly persistent trends on regime 1 with probability values 0.96 and 0.84, respectively as compared to CNY, KRW and USD, while the latter have high persistent trends on regime 2 with probability values, 0.99, 0.95, 0.82, respectively.

MULTIFRACTALITY AND LONG-RANGE DEPENDENCE OF ASSET RETURNS: THE SCALING BEHAVIOR OF THE MARKOV-SWITCHING MULTIFRACTAL MODEL WITH LOGNORMAL VOLATILITY COMPONENTS

2008 ◽  
Vol 11 (05) ◽  
pp. 669-684 ◽  
Author(s):  
RUIPENG LIU ◽  
T. DI MATTEO ◽  
THOMAS LUX
Keyword(s):  
Exchange Rates ◽  
Foreign Exchange ◽  
Empirical Data ◽  
Long Memory ◽  
Markov Switching ◽  
Asset Returns ◽  
Financial Data ◽  
Long Range Dependence ◽  
Foreign Exchange Rates ◽  
Good Agreement

In this paper, we consider daily financial data from various sources (stock market indices, foreign exchange rates and bonds) and analyze their multiscaling properties by estimating the parameters of a Markov-switching multifractal (MSM) model with Lognormal volatility components. In order to see how well estimated models capture the temporal dependency of the empirical data, we estimate and compare (generalized) Hurst exponents for both empirical data and simulated MSM models. In general, the Lognormal MSM models generate "apparent" long memory in good agreement with empirical scaling provided that one uses sufficiently many volatility components. In comparison with a Binomial MSM specification [11], results are almost identical. This suggests that a parsimonious discrete specification is flexible enough and the gain from adopting the continuous Lognormal distribution is very limited.

Sign in / Sign up

Export Citation Format

Share Document