Back

PROJECTS

• Master Thesis (Econ.)
• Master Thesis (Theo. Physics)
• Bachelor Thesis
• Teaching Experience
• Consulting
• Specialty coffee shop recommendations

• Back

MASTER THESIS (Economics)

Are women less likely to renegotiate? A threshold model of the wage renegotiation decision

Keywords: Gender economics, labour economics, wage renegotiation, gender pay gap, SOEP

This thesis employed a structural approach to examine whether women are less likely to renegotiate. We developed a threshold model of the salary renegotiation decision process based on risk profile and wage dissatisfaction to investigate gender differences in the propensity to ask for higher wage compensation. We used data from the German Socio-Economic Panel (SOEP-Core v36) for estimation purposes and constrained our sample to match our renegotiation context. Our theoretical framework yields a latent variable setup, the additive random utility model with type 1 extreme value distribution. The two unobserved parameters - renegotiation success probability p and cost c - were retrieved in parameter constrained logit regressions in STATA. Focusing on gender differences in initiating salary renegotiations, we uncovered p and c separately for men and women and estimated the gender difference in the coefficients.Having thus specified all the components of our theoretical framework, we can use this information to 1) predict renegotiations and 2) examine what would happen if women were endowed with the male parameters (p_m or c_m). We visualised the predicted engagement in renegotiations in heat maps created in Python.The gender difference in triggering wage renegotiations is significant. Men have a higher success probability than women but face higher renegotiation costs. Women are predicted to be less likely to ask for a higher wage compensation. However, the predicted female engagement in wage bargaining considerably increases when assigned the male parameters p_m or c_m. The significant gender difference in the risk aversion distribution drives these results.

• Back

MASTER THESIS (Theoretical Physics)

A model for emergent behaviour in a competitive market

Keywords: Econophysics, Game Theory, boom-bust cycles, commercial centres

Econophysics employs statistical physics methods to analyse economic problems such as marketplace interactions. Despite its simplicity, the marketplace model of competing sellers proposed by Mitchell and Ackland displays a rich behaviour: An asynchronous update scheme leads to price oscillations of irregular periods. These boom-bust cycles are asymmetric and consist of rapid upswings and subsequent slower declines.My thesis extended the simple marketplace model to allow emergent variations in the local population: previously passive buyers can now move away from regions with no vendors. Three different relocation principles - randomness, proximity and price - have been implemented, compared to the original model, and their impact on the emergence of commercial centres has been investigated.Modelling the choice of the system’s four free parameters - ∆,γ, mode and β - as an extensive form game and applying backward induction, leads to the optimal parameter set ((∆, γ), (mode, β)) for the three phase space regimes:
• ((0.2, 0.1), (local, 0.1)) for the all-price regime,
• ((0.02, 0.55), (empty, 0.1)) for the oscillatory regime and
• ((0.02, 0.8), (empty, 0.5)) for the Bertrand regime.Programming was done in Python.

• Back

BACHELOR THESIS

Preventing the extremism propagation in the Deffuant-Weisbuch model of Opinion Dynamics via introduction of "counter-extremist" agents

Keywords: Physics of socio-economic systems, Computational Physics, Opinion Dynamics, Deffuant-Weisbuch model, extremism propagation, counter-extremist agents, Utopia framework

Several publications regarding the Deffuant-Weisbuch model - one of the most important models in opinion dynamics - have shown that already a small fraction of so-called “extremist agents“, i.e. agents with extreme opinions and low uncertainties, can dominate the opinion dynamics process.My thesis investigated whether introducing a third group of agents - the so-called “counter-extremist agents“ - can reverse the observed “extremism propagation“ in the Deffuant-Weisbuch model by exercising a) social control via coercion or the force of the public opinion or b) the authority to influence (here: as experts). Counter-extremist agents are based on three of French and Raven's bases of social power: expert power, coercive power and referent power.I recoded and extended the Deffuant-Weisbuch model of opinion dynamics in the Utopia modeling framework developed by my workgroup. Programming was done in C++ and the subsequent data analysis in Python.The main findings were that all three types of “counter-extremist agents“ can prevent the extremism propagation - the force of the public opinion (referent power) being the most effective. Furthermore, for a susceptibility of 0.5, the experts were more potent than the coercion, whereas the opposite held for a high susceptibility of 0.8.

• Back

TEACHING EXPERIENCE

• Back

QUICK COMPILATION OF RECOMMENDED SPECIALTY COFFEE SHOPS (in alphabetical order)

Bratislava
Good Times coffee roastersBudapest
CAUSE Café & Roastery, My Little Melbourne CoffeeCambridge
Bould Brothers CoffeeCopenhagen
April, La Cabra, The ArtisanDublin
3fe, Cloud Picker, KaphEdinburgh
Format, Fortitude Coffee, Hideout, Lowdown Coffee, The Milkman, Union Brew Lab
wider Edinburgh area: Unorthodox Roasters (Kinross)Glasgow
Laboratorio EspressoHeidelberg
coffee nerd, Florian Steiner Kaffeerösterei, KaffeezimmerLisbon
Copenhagen Coffee Lab, Espressolab, Fábrica Coffee Roasters, The CoffeeNYC
Black Fox Coffee, La Colombe Coffee Roasters, Suited, % ArabicaReykjavik
Reykjavik RoastersSplit
D16, Stow Coffee RoastersValletta
Lot 61 Coffee RoastersVienna
Caffè Couture, GOTA Coffee Experts, Jonas Reindl, Kaffein, The Truth Wollzeile

• Back

CONSULTING

Project Work

Mentorship schemes

Talent programmes