Kamila Nowakowicz

Kamila Nowakowicz

Job Market Candidate

Department of Economics

Connect with me

Languages
English, Polish
Key Expertise
Econometric Theory

About me

Kamila is a PhD candidate in the Department of Economics. She is on the job market in 2024/25. Her research focuses on econometric theory. She gained both her BSc and MSc in Econometrics and Mathematical Economics at the LSE.

In her job market paper, she proposes a nonparametric bootstrap procedure for network data. Her other working papers develop methods for testing for a wide range of shape restrictions and additivity.

Contact Information

Email
k.nowakowicz@lse.ac.uk

Office Address
Department of Economics
London School of Economics and Political Science
Houghton Street, London WC2A 2AE

Contacts and Referees

Placement Officer
Matthias Doepke

Supervisors
Javier Hidalgo
Taisuke Otsu

References
Javier Hidalgo
Department of Economics
London School of Economics and Political Sciences
Houghton St, London WC2A 2AE
f.j.hidalgo@lse.ac.uk

Tatiana Komarova
Faculty of Economics
Austin Robinson Building
Sidgwick Avenue, Cambridge CB3 9DD
tk670@cam.ac.uk

Taisuke Otsu
Department of Economics
London School of Economics and Political Sciences
Houghton St, London WC2A 2AE
t.otsu@lse.ac.uk

 

Download CV

Job Market Paper

Nonparametric network bootstrap
We propose a bootstrap procedure for network data based on a nonparametric linking function estimator. We characterise when the nonparametric linking function estimator is uniformly consistent. We provide conditions under which a class of functions related to U-statistics on the bootstrapped networks converge weakly in probability to the same limiting distribution as the corresponding statistics on the original network. We prove bootstrap consistency in the sense that a Wasserstein distance between the bootstrap network generating distribution and the true network generating distribution goes to zero in probability. Monte Carlo simulations show good confidence interval coverage for a wider class of network functions than are accounted for by our theory. An application to the data from Banerjee, Chandrasekhar, Duflo, and Jackson (2013) not only replicates their findings, but also shows that our method can be applied under weaker assumptions and smaller sample sizes. We propose an alternative specification of their model which takes advantage of our linking function estimator and may be of interest independently of our bootstrap procedure. I Link to paper.

Publications and Research

Working Papers

Understanding regression shape changes through nonparametric testing, with Tatiana Komarova. 
We propose a procedure for testing whether a nonparametric regression mean satisfies a shape restriction that varies within the domain of the regressor. Notably, the change points of these shape restrictions are unknown and must be estimated. Our test statistic is based on the empirical process, drawing inspiration from Khmaladze (1982). This paper extends the nonparametric methodology of Komarova and Hidalgo (2023) by proposing a method to estimate the shape change points and consequently addressing the additional estimation errors introduced by that stage. We analyse strategies for managing these errors and adapting the testing approach accordingly. Our framework accommodates various common shapes, such as (inverse) U-shapes, S-shapes, and W-shapes. Furthermore, our method is applicable to partial linear models, thereby encompassing a broad spectrum of applications. We demonstrate the efficacy of our approach through application to several economic problems and data.

Testing for additivity in nonparametric regression models, with Javier Hidalgo and Tatiana Komarova. 
We describe and examine a test for additivity in a nonparametric framework using partial sums empirical processes.