New Master/Ph.D. Students and Postdoctoral Researcher to join “Big Data, Artificial Intelligence, and Machine Learning in Finance in Turkey”

We are looking to admit four (4) outstanding applicants to the “Big Data, Artificial Intelligence, and Machine Learning in Finance in Turkey” project at the Department of Economics, Middle East Technical University, Ankara, Turkey. Highly motivated (Turkish or international) Master/Ph.D. students (3 positions) and Postdoctoral researcher (1 position) are encouraged to join our recent research project funded by “TÜBİTAK 2232 programme” emphasize on “Big Data, Artificial Intelligence, and Machine Learning in Finance in Turkey”.

Project Summary

Nowadays, Big Data affects every scientist in any field, from Astronomy to Zoology. Our approach integrates state-of-the-art concepts and methods from distinct fields of knowledge, Economics/Finance/Econometrics and Network Sciences (an area that draws theory and methods from computer science, data science, and statistics) with Artificial Intelligence (AI), thus creating a new integrative methodological space. We blend the AI and network theory with empirical approaches such as Hawkes Process (HP) and econometrics with a ML perspective, as it gets beyond correlation to causality and prediction. We aim to design structural content for information disclosure, financial efficiency, sentiment analyses and macroeconomic forecasting and nowcasting in Turkey. Our transformative idea merges data science with AI and ML, Natural Language Processing (NLP) techniques and using these insights, we not only utilize deep learning to exploit networks within the economy and the financial sector in Turkey but also create methods to trace a range of data sources which matter for quantifying the endogenous feedback mechanism. Hence, the proposed research has an innovative approach in Turkey to establish a multidisciplinary environment between economists, data scientists, and computer scientists.

Project Objectives

This project plans to create, establish and manage a wealth of new data sources that moves financial institutions into the realm of big data, as well as model and understand the statistical nature of this computationally demanding unstructured data. We will extract relevant financial information for sentiment analysis through NLP; describe the co-existence, and the interplay between the exogenous impact of news and the endogenous price mechanism, over a feedback loop, utilizing the Hawkes model. Our second objective is to apply the ML approach to time series setting and present a modeling framework to forecast and nowcast the main macroeconomic indicators in Turkey on a medium-term horizon of two years. We will use training and test framework, feed-forward artificial neural network, and Shapley decomposition for an econometric ML approach.

Our approach is transformative, as it requires the integration of concepts and methodologies from distinct fields of knowledge. In doing so, the final objective of our research is to further interdisciplinary collaborations among several scientific experts