Forecasting of the Unemployment Rate in Turkey: Comparison of the Machine Learning Models

Mehmet Güler (), Ayşıl Kabakçı, Ömer Koç, Ersin Eraslan, K. Hakan Derin, Mustafa Güler (), Ramazan Ünlü, Yusuf Sait Türkan and Ersin Namlı
Additional contact information
Mehmet Güler: Department of Labor Economics and Industrial Relations, Faculty of Economy, Istanbul University, Istanbul 34452, Turkey
Ayşıl Kabakçı: Department of Labor Economics and Industrial Relations, Faculty of Economy, Karadeniz Technical University, Trabzon 61080, Turkey
Ömer Koç: Department of Labor Economics and Industrial Relations, Faculty of Economy, Istanbul University, Istanbul 34452, Turkey
Ersin Eraslan: Department of Property Protection and Security, Vocational School of Social Sciences, Niğde Ömer Halis Demir University, Niğde 51100, Turkey
K. Hakan Derin: Department of Property Protection and Security, Vocational School of Golbasi, Adiyaman University, Adiyaman 02500, Turkey
Mustafa Güler: Engineering Faculty, Department of Engineering Sciences, Istanbul University-Cerrahpaşa, Avcılar 34320, Turkey
Ramazan Ünlü: Engineering Faculty, Department of Industrial Engineering, Abdullah Gul University, Kocasinan 38125, Turkey
Yusuf Sait Türkan: Engineering Faculty, Department of Industrial Engineering, Istanbul University-Cerrahpaşa, Avcılar 34320, Turkey
Ersin Namlı: Engineering Faculty, Department of Industrial Engineering, Istanbul University-Cerrahpaşa, Avcılar 34320, Turkey

Sustainability, 2024, vol. 16, issue 15, 1-17

Abstract: Unemployment is the most important problem that countries need to solve in their economic development plans. The uncontrolled growth and unpredictability of unemployment are some of the biggest obstacles to economic development. Considering the benefits of technology to human life, the use of artificial intelligence is extremely important for a stable economic policy. This study aims to use machine learning methods to forecast unemployment rates in Turkey on a monthly basis. For this purpose, two different models are created. In the first model, monthly unemployment data obtained from TURKSTAT for the period between 2005 and 2023 are trained with Artificial Neural Networks (ANN) and Support Vector Machine (SVM) algorithms. The second model, which includes additional economic parameters such as inflation, exchange rate, and labor force data, is modeled with the XGBoost algorithm in addition to ANN and SVM models. The forecasting performance of both models is evaluated using various performance metrics such as Mean Absolute Error (MAE), Root Mean Square Error (RMSE), and Mean Absolute Percentage Error (MAPE). The findings of the study show how successful artificial intelligence methods are in forecasting economic developments and that these methods can be used in macroeconomic studies. They also highlight the effects of economic parameters such as exchange rates, inflation, and labor force on unemployment and reveal the potential of these methods to support economic decisions. As a result, this study shows that modeling and forecasting different parameter values during periods of economic uncertainty are possible with artificial intelligence technology.

Keywords: unemployment; machine learning; sustainable economy; labor market (search for similar items in EconPapers)
JEL-codes: O13 Q Q0 Q2 Q3 Q5 Q56 (search for similar items in EconPapers)
Date: 2024
References: View references in EconPapers View complete reference list from CitEc
Citations:

Downloads: (external link)
https://www.mdpi.com/2071-1050/16/15/6509/pdf (application/pdf)
https://www.mdpi.com/2071-1050/16/15/6509/ (text/html)

Related works:
This item may be available elsewhere in EconPapers: Search for items with the same title.

Export reference: BibTeX RIS (EndNote, ProCite, RefMan) HTML/Text

Persistent link: https://EconPapers.repec.org/RePEc:gam:jsusta:v:16:y:2024:i:15:p:6509-:d:1446076

Access Statistics for this article

Sustainability is currently edited by Ms. Alexandra Wu

More articles in Sustainability from MDPI
Bibliographic data for series maintained by MDPI Indexing Manager ().