Doruk Gokalp
PhD Economics Candidate, Data Scientist
Profile
NYU Economics PhD candidate specializing in forecasting, time-series econometrics, statistical learning, and
large-scale quantitative modeling. Experience developing forecasting models and analytical tools at Google using
Python and SQL, with additional expertise in Bayesian methods, structural modeling, and HPC.
Experience
Data Scientist Research Intern
Google
- Developed a dynamic survival analysis framework to model non-linear hazard decay in long-horizon subscriber retention, reducing multi-year forecasting error by up to 10x across subscription tiers.
- Developed a structural model of subscription growth using empirical survival durations. Built a reusable implementation library and maintained stakeholder dashboards. Authored a reusable skill for internal agentic coding tools to automate dashboard production.
May 2026 – Aug 2026
Lecturer and Teaching Assistant
NYU, UPF, BSE, METU
- Lecturer for undergraduate Econometrics across four terms (2023–2025) and Statistics (Spring 2026).
- Teaching Assistant for Graduate Macroeconomics across NYU MA and MSQE programs.
2020 – Present
Research Assistant
NYU
- Built high-frequency forecasting and shock-decomposition models for asset price dynamics and macro news, integrating event-driven features into predictive pipelines.
- Scaled estimation workflows on HPC clusters using vectorization and parallelization to support large predictive workloads and rapid iteration.
Jan 2023 – Jan 2025
Research Assistant
Center for Research in International Economics (CREi)
- Developed and simulated macroeconomic models to forecast trade and policy shocks; executed counterfactual scenarios to quantify impacts. Delivered quantitative projections and reproducible MATLAB/Stata code that supported academic and policy analysis.
Apr 2020 – Sep 2021
Education
New York University
PhD in Economics
2021 – 2027 (Expected)
Universitat Pompeu Fabra
Master of Research in Economics
2020 – 2021
Barcelona School of Economics
M.Sc. in Economics and Finance
2019 – 2020
Ranked 1st in cohort
Middle East Technical University
B.Sc. in Mathematics and Economics
2014 – 2019
Dual Degree, High Honor
Technical Skills
Languages
Python
SQL
R
MATLAB
C++
Methods
Machine learning
Bayesian estimation
Causal inference
Structural modeling and estimation
Time-series forecasting
Panel and high-frequency econometrics
State-space and Markov switching models
Tools
AI-assisted development (Codex, Antigravity)
Git
Stata; distributed & high-performance computing
Research
These abstracts present research conducted independently and jointly with the co-authors listed.
Full paper drafts are available upon request.
Gokalp, D. (2026)
‘One Big Beautiful Tax Shield, The Macroeconomics of Permanent Full Expensing’
‘One Big Beautiful Tax Shield, The Macroeconomics of Permanent Full Expensing’
The One Big Beautiful Bill Act (OBBBA) of 2025 marks a structural shift in U.S. tax policy by making 100\% bonus depreciation, the immediate expensing of qualifying investment, a permanent feature of the tax code rather than a temporary, countercyclical stimulus device. Under perfect capital markets, the effects of this policy are confined to changes in the user cost of capital. With financial frictions, however, the policy primarily operates through a liquidity channel. I quantify these effects in a recursive dynamic equilibrium model of heterogeneous firms with financing frictions and lumpy investment. The model measures long-run effects on capital and output, characterizes the transition from a temporary to a permanent regime, and assesses how permanent expensing changes cyclical propagation of aggregate shocks. Ultimately, the analysis asks whether efficiency gains from relaxing financial frictions justify the fiscal costs and windfall transfers associated with permanently embedding full expensing into the tax code.
Working Paper
Gokalp, D., Ventura, J., Yesilbayraktar, U. (2026)
‘The United States of Europe’
‘The United States of Europe’
We quantify the economic and political effects of removing internal borders within Europe. We develop a spatial model of trade and public service provision in which governments redistribute income, deliver public goods whose effectiveness depends on political alignment, and face border‐related trade frictions. Political alignment is measured from vote‐share‐weighted party positions, yielding a region’s political distance from its central government. The model is calibrated to European regional data on incomes, populations, trade flows, and political preferences. Counterfactual integration eliminates all internal borders, allowing us to decompose welfare changes into contributions from redistribution, trade cost reductions, political alignment, and shifts in the public spending base. The analysis reveals substantial heterogeneity in the regional gains and losses from integration, reflecting spatial differences in both economic and political geography.
Working Paper
Gokalp, D. (2025),
‘Dynamics and Distributional Consequences of Biased Household Inflation Expectations in Heterogeneous Agents Economies’
‘Dynamics and Distributional Consequences of Biased Household Inflation Expectations in Heterogeneous Agents Economies’
This paper documents systematic differences in the dynamics of household inflation expectations when households are categorized by income and education levels. Assuming that households continuously learn an inflation trend, I illustrate that households from different socioeconomic strata adjust their long-term beliefs at varying rates, and their short-term inflation forecasts result from differing perceptions of the persistence of the current inflation level. The paper further documents the steady-state effects of persistent inflation biases within a heterogeneous agents model in partial equilibrium. Quantitative solutions of the model under realistic calibrations reveal that even minor differences in inflation expectations can lead to significant nominal wealth inequality in the long run for households with biased expectations.
Working Paper
