Show/Hide Menu
Hide/Show Apps
Logout
Türkçe
Türkçe
Search
Search
Login
Login
OpenMETU
OpenMETU
About
About
Open Science Policy
Open Science Policy
Open Access Guideline
Open Access Guideline
Postgraduate Thesis Guideline
Postgraduate Thesis Guideline
Communities & Collections
Communities & Collections
Help
Help
Frequently Asked Questions
Frequently Asked Questions
Guides
Guides
Thesis submission
Thesis submission
MS without thesis term project submission
MS without thesis term project submission
Publication submission with DOI
Publication submission with DOI
Publication submission
Publication submission
Supporting Information
Supporting Information
General Information
General Information
Copyright, Embargo and License
Copyright, Embargo and License
Contact us
Contact us
A comparative study of classical and machine learning approaches for time series forecasting: an empirical analysis on exports in Turkey
Download
index.pdf
Date
2019
Author
Günel, Eda
Metadata
Show full item record
This work is licensed under a
Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License
.
Item Usage Stats
524
views
292
downloads
Cite This
Exports has become one of the main economic indicator for countries. Accordingly, an accurate forecasting for exports is an important step for decision making and finding the most appropriate forecasting model constitutes the main subject of many studies. By taking the popularity and success of the machine learning (ML) methods on time series forecasting tasks into consideration, they are utilized also in this study to observe their predictive performances on Turkish exports. In this respect, Long Short Term Memory (LSTM), Support Vector Machines (SVM) and Random Forest (RF) are applied and the results are compared with the most commonly used classical time series models such as Autoregressive Integrated Moving Average (ARIMA) and Exponential Smoothing (ETS) models. The analysis is conducted on Turkish monthly exports data taken from Turkish Statistical Institute (TURKSTAT) within the time interval of January 1997 – September 2019 and the main steps of the analysis are anomaly detection and cleaning, data preprocessing, model development, hyperparameter tuning and model selection and model comparison. The main findings can be summarized as follows; the anomaly detection and cleaning process improves the forecasting ability of the models, ETS is the best forecasting model and SVM model v is the most promising among the ML models and the most competitive with the leading one. Besides, ARIMA has the poorest generalization ability among the others.
Subject Keywords
Machine learning
,
Keywords: machine learning
,
time series
,
random forest
,
long short term memory
,
support vector machine
,
exports forecasts
URI
http://etd.lib.metu.edu.tr/upload/12625526/index.pdf
https://hdl.handle.net/11511/45763
Collections
Graduate School of Natural and Applied Sciences, Thesis
Suggestions
OpenMETU
Core
Analysis of economic data using robust methods
Coşar, Evren Erdogan; Akkaya, Ayşen; Department of Statistics (2002)
Export plays an important role in the economic development of a country. Because of its importance in implementing trade policies, there is an extensive literature on the estimation of export functions. Most of the studies on estimating export functions utilized conventional econometric techniques with the normality assumption of innovations. However, it has been recently recognized that non-normal distributions for innovations are more common in practice. IllThe main aim of this study is to estimate the ex...
Analyses of Azerbaijan’s foreign trade by integrating volatilities of rates of changes in oil prices and exchange rates
Suleymanli, Shohrat; Gaygısız Lajunen, Esma; Department of Economics (2022-12)
Oil price and real effective exchange rate, which are the determining factors for export- import and Gross Domestic Product (GDP), have been the subject of much research in oil-exporting countries. Because of Azerbaijan’s dependence on crude oil export, the fluctuations in oil prices, together with exchange rate movements, have implications for the Azerbaijanian economy. This research examines (a) the relationships between export-import growth and GDP growth, (b) impacts of rates of changes in oil price and...
A Life cycle assessment based decision support tool for early-design phase of mass-housing neighbourhoods in Turkey
Kayaçetin, Nuri Cihan; Tanyer, Ali Murat; Department of Building Science in Architecture (2018)
In developing countries such as Turkey, the main driving force of economy is the Architecture-Engineering-Construction (AEC) industry. Hence, rapid urban expansion dramatically increases the pressure on the existing infrastructure which affects buildings, public transportation and overall energy usage. In order to control urbanization, governments facilitate mass-housing projects in increasing numbers. Higher rates in construction require an immediate need for methods of assessing the environmental impacts ...
The Investigation of recent trends in Turkish current account
Yüksel, Gökay; Cömert, Hasan; Department of Economics (2015)
International trade is still indispensable for each country in today’s world for both economic and political reasons. Correspondingly, world trade volume has been uninterruptedly rising. By the same token, Turkish trade volume, in line with other developing countries, has been also growing. However, Turkish imports have been rising much faster than Turkish exports since the 1990s. As a result, Turkey has been chronically suffering from current account deficit problem for years. What is more, Turkish current...
The investigation of recent trends in Turkish current account
Cömert, Hasan; Gökay, Yüksel (2015-11-01)
International trade has been very important for both economic and political reasons. Correspondingly, world trade volume has been uninterruptedly rising. By the same token, Turkish trade volume, in line with other developing countries, has been also growing. However, Turkey has been chronically suffering from current account deficit problems for years. The main objective of the article is to search for the determinants of Turkish current account and a possible structural break in it. There are two main find...
Citation Formats
IEEE
ACM
APA
CHICAGO
MLA
BibTeX
E. Günel, “ A comparative study of classical and machine learning approaches for time series forecasting: an empirical analysis on exports in Turkey,” Thesis (M.S.) -- Graduate School of Natural and Applied Sciences. Statistics., Middle East Technical University, 2019.