We develop a hybrid model of multivariate volatility that uses Recurrent Neural Networks to capture the conditional variances of latent orthogonal factors in a GO-GARCH framework. Our approach seeks to balance model flexibility with ease of estimation and can be used to model conditional covariances of a large number of assets. The model performs favourably in comparison with relevant benchmark models in a Minimum Variance Portfolio (MVP) scenario.

Economic inequality has been found to be associated with increased unethical behavior and an increased acceptance of unethical behavior. In this paper we explored whether higher amounts of perceived inequality lead to an increase in the expectation of unethical behavior. We tested whether people would say that they themselves would engage in more unethical behavior in a context of high compared to low inequality. We find evidence for this hypothesis in 3 of 4 studies (n = 3,038). An internal meta-analysis shows a small but significant effect. Such increased expectations that oneself will behave unethically likely has consequences for societal trust and functioning.

This paper extends binary outcome models by introducing a threshold latent variable with an unobserved and serially correlated factor structure. I derive conditions for point identification and sharp set identification under relatively weak model assumptions. I further provide consistency and unimodality proofs for a proposed simulated maximum likelihood estimator, employing a novel EM algorithm. Monte Carlo simulations demonstrate favorable MSE and bias properties in finite samples. An application to Klette & Kortum (2004) R&D innovation highlights a viable structural interpretation of the econometric theory of factors.

Presentations / Posters: University of Toronto (Econometrics Seminar), CIREQ Econometrics Meeting 2025, Canadian Economic Association Conference 2025 , RCEA International Conference 2025, Celebrating James MacKinnon Conference, 40th CESG Annual Meeting, Causal Inference Across Fields 2025 (Data Science Institute), SNDE Young Researchers Workshop 2025

A novel kernel representation of economic models using Markov chain theory (Strassen 1965) is developed. This approach builds on, and is yet distinct from, the model verification and optimal transport literature (Ekeland et al 2010). Multiple equilibria (Jovanovic, 1989), model incompleteness (Tamer2003), and partially identified counterfactuals (Chesher and Rosen, 2021) can be captured by a correspondence from observed or latent variables into this kernel space. Using a stochastic line representation from integral geometry (Santalo, 2004), analytical sharp sets are derived, which importantly apply to cross-sections as well as panels. Empirically, Dirichlet process methods improve upon Artstein's (1983) inequalities in both frequentist (Beresteanu et al., 2012) and Bayesian (Kline and Tamer, 2016) inference. Estimation of production functions, propensity scores, entry decisions, and hidden Markov models illustrate that bringing nonparametric kernels to the data allows for a much richer investigation of multiple equilibria, learning, and uncertainty regimes.

Binary outcome models in panels often involve fixed effects in the presence of serial- as well as cross-sectional dependence (Arellano et al., 2007). To this end, we extend a jackknife correction method from the literature regarding dynamics of observed outcomes  yi,t-1 (Pakel, 2019) to a class of normal models involving latent dynamics in y* i,t-1. We provide local and global identification results. Estimation extends marginal and pairwise composite likelihood methods (Tuzcuoglu, 2023; Cox and Reid, 2004) to cross-sectionally dependent data. Simulations display well behaved finite sample properties and inference procedures. Leveraging the dataset of Laeven et al, 2013, we apply our methodology to analyze systemic banking crises of around 100 countries across 40 years. Results show the significant determinants of banking crises. Moreover, interpreting the time fixed effects as the global systemic financial risk, we document how it evolves over time and its relation with other well-known risk indices.

This paper examines the impact of exogenous oil supply shocks on dairy production under the Canadian Supply Management system. Effects are compared to impacts of observed cost shifters. Substitution of goods produced under the quota system occurs in response to the shock. Impulse response functions, long-run effects, and short-term impacts are discussed.