Modeling of exchange rates by multivariate adaptive regression splines and comparison with classical statistical methods

Download
index.pdf
2017
Köksal, Ece
Show full item record
363
views
134
downloads
Economic factors like inflation, interest rates and exchange rates are among the leading indicators of a country’s relative level of economic health. With the help of technological improvements and global requirements, trading volume and a wide range of commerce network, exchange rates play a vital role in economics and finance since a higher exchange rate may result in a lower trade balance of a country, whereas a lower rate may cause an increase. Inflation, interest rates, domestic money supply growth, a country’s balance of payments’ size and trend, a country’s economic growth, dependency on outside sources and central bank intervention, are the factors which affect an exchange rate. Since many dependent and independent factors affect exchange rates, it is difficult to predict them. In areas of application, data mining is frequently used for decision support, financial forecasting, marketing strategy, prediction, etc. The method of data mining and machine learning is applied to analyze and forecast the future behavior of such complex systems. Modeling and prediction of exchange rates are still a challenge, although mathematicians, economists and statisticians have worked to reach a model with a superior forecasting ability for many years. Therefore, in this study, we aim to generate mathematical models to forecast the monthly US Dollar (USD) / Turkish Lira (TRY) and Euro (EUR) / Turkish Lira (TRY) exchange rates via data mining tools. For this purpose, we apply a flexible model Multivariate Adaptive Regression Splines (MARS) and widely used models Linear Regression (LR) and Support Vector Regression (SVR). In this study, MARS, LR and SVR models applied on USD / TRY and EUR / TRY exchange rate data sets in the period of 01/01/2007 and 30/04/2015; then the results of these models are compared and found out that MARS method has superior forecasting ability over LR and SVR methods for USD / TRY and EUR / TRY exchange rates. The thesis ends with a conclusion and an outlook to future investigations. 
Foreign exchange., Regression analysis ., Mathematical statistics., Foreign exchange rates., Foreign exchange market.
http://etd.lib.metu.edu.tr/upload/12621053/index.pdf
https://hdl.handle.net/11511/26705
Graduate School of Applied Mathematics, Thesis

Suggestions

Exchange rate pass-through into domestic price indicators: a sectoral analysis of Turkish economy
Özen, Emine Özgü; Akbostancı Özkazanç, Elif; Department of Economics (2011)
The question of exchange rate pass-through into domestic inflation is a widely analyzed issue due to its importance as regards to monetary policy, exchange rate policy and in general macroeconomic policy for open economies. Although most of the literature is focused on the exchange rate pass-through at the aggregate level, there are fewer studies that are done at the sectoral level for the Turkish economy. In this study by using a distribution chain of pricing model developed by McCarthy (2000), pass-throug...
Exchange rate pass-through to domestic prices in Turkish economy
Alper, Koray; Gaygısız Lajunen, Esma; Department of Economics (2003)
In this study, determinants and the evolution of the exchange rate passthrough to domestic inflation in the Turkish economy is analyzed. The analyses cover the 1987-2003 period. In the analyses, single equation أError Correction Modelsؤ are used to estimate the exchange rate pass-through. Estimation results suggest that alike other emerging countries, the degree of exchange rate passthrough to domestic prices is high and the pass-through is completed in a very short time span. Estimations results also indic...
The Possibility of financial crisis in Developing Countries under flexible exchange rate regimes : a multidimensional approach
Çolak, Mehmet Selman; Cömert, Hasan; Department of Economics (2012)
Many economists and politicians have blamed fixed exchange rate regimes for several crises taking place in developing countries after the 1980s. According to them, since the beginning of the 2000s, widespread implementation of flexible exchange rate regimes and high international reserves have prevented developing countries from experiencing similar catastrophic experiences. This interpretation seems to be misleading. We believe that even flexible exchange rate regimes with high international reserves do no...
The Effects of exchange rates, oil prices, global risk perceptions and global warming on food prices
Dağdelen, Derya; Soytaş, Uğur; Department of Business Administration (2012)
This thesis examines the relationship between food prices, oil prices, carbon emission prices, exchange rates and global risk perception. To obtain the effects of these variables on the food prices, Toda and Yamamoto procedure is employed for 5-day week daily time series covering the period February 27, 2008 and March 21, 2011. The empirical results indicate that only volatility index Granger causes food prices. Furthermore, according to results of generalized impulse response plots food prices respond to a...
Testing the long-run validity of purchasing power parity for Eastern and Southern Europe Countries
Neslihan, Can; Yıldırım Kasap, Dilem; Department of Economics (2013)
This study examines the long-run validity of the Purchasing Power Parity, namely stationarity of real exchange rates for six Eastern and Southern Europe countries over the period of January 1994- December 2012. Despite extensive research to date, the empirical evidence on the PPP hypothesis remains inconclusive. A number of very recent studies ascribe this to the low power of the standard unit root tests in identifying stationarity of potentially nonlinear real exchange rates. In this sense, this study empl...
Citation Formats
E. Köksal, “Modeling of exchange rates by multivariate adaptive regression splines and comparison with classical statistical methods,” M.S. - Master of Science, Middle East Technical University, 2017.
METU IIS - Integrated Information Service open@metu.edu.tr