I’m Baris Soybilgen, an experienced data scientist and economist based in Stockholm. With over a decade of experience in the field, I’ve sharpened my skills in predictive modeling, statistical evaluation, and data visualization. My academic background, which includes a Ph.D. in Economics and a Bachelor of Science in Systems Engineering, has enhanced my analytical capabilities.
I’ve had the privilege to work with diverse organizations, including the Sambla Group in Stockholm, where I currently contribute as a Data Scientist. Prior to this, I’ve worked as a data analyst at The Governance and Local Development Institute in Gothenburg and assistant professor at Istanbul Bilgi University in Istanbul. Throughout my journey, I’ve been driven by the passion to leverage data in innovative ways that yield impactful results.
My technical proficiency spans across multiple platforms and languages, including Python, R, Matlab, and SQL. Apart from these, I’m adept at developing cloud applications and creating interactive dashboards.
In addition to my professional pursuits, I am conducting several research projects such as developing food price indices by scraping web-based food price data daily; predicting GDP growth rates with tree-based ensemble machine learning models and dynamic factor models; identifying business cycle states using neural networks.
Thank you for visiting my personal space. I look forward to connecting with like-minded professionals and exploring collaborative opportunities!
This study uses a sample of daily food prices scraped from retail chains’ websites for the period from July 2018 to December 2020, comprising over 5.9 million data points. Using these food prices, we construct 132 food price subindexes compatible with official data published by the Turkish Statistical Institute (Turkstat), which are published only once a month. We then use the online food price subindexes to calculate the primary food inflation rate. Our results show that the online index successfully nowcasts the official inflation rates, providing results considerably earlier than the official rate is announced.
In this study, we nowcast quarter-over-quarter US GDP growth rates between 2000Q2 and 2018Q4 using tree-based ensemble machine learning models, namely, bagged decision trees, random forests, and stochastic gradient tree boosting. For obtaining variables from our large-scale data set, we adopt a dynamic factor model. Our results show that tree-based ensemble models usually outperform linear dynamic factor models. Factors obtained from real variables appear to be more influential in machine learning models.
We use dynamic factors and neural network models to identify current and past states (instead of future) of the US business cycle. In the first step, we reduce noise in data by using a moving average filter. Then, dynamic factors are extracted from a large-scale data set consisted of more than 100 variables. In the last step, these dynamic factors are fed into the neural network model for predicting business cycle regimes. We show that our proposed method follows US business cycle regimes quite accurately in sample and out of sample.
In this study, we propose a flexible two-stage least square regression to deal with both instability and endogeneity problems in the policy reaction function of the Central Bank of the Republic of Turkey (CBRT). By analyzing the period between 2006 and 2019, we clearly show that the monetary policy function of the CBRT changes over time, and using a time-invariant monetary policy rule model would yield misleading results.
The paper analyzes the point and density predictive performance of alternative nowcast combination schemes in the context of bridge equations for the Turkish unemployment rate. Furthermore, we also nowcast the unemployment rate by using dynamic factor models (DFMs). Our results indicate that most of the sophisticated forecast combination methods have better predictive accuracy than the simple forecast combinations, especially in higher forecast horizons, which constitutes a case for the nowcast combination puzzle.
Real gross domestic product (GDP) data in Turkey are released with a very long delay. This means that policy makers, the media, and market practitioners have to infer the current state of the economy by examining data that are more timely and are released at higher frequencies than the GDP. This paper proposes an econometric model that allows us to read through these more current and higher-frequency data automatically, and translate them into nowcasts for the Turkish real GDP. Our model outperforms the nowcasts produced by the Central Bank of Turkey, the International Monetary Fund, and the Organisation for Economic Co-operation and Development.