A seminar by Professor Shangjin Wei from Columbia Business School

Title: Why Financial Sanctions? The Case of the SWIFT Ban and Western Bank Withdrawal on Russian Trade

Abstract: Since trade sanctions can already restrict the trade of targeted countries or firms, why do we also need financial sanctions? With highly disaggregated transaction level Russian trade data, we find that trade sanctions do not affect Russian trade with non-Western countries while they are effective in reducing Russian trade with Western countries. In contrast, financial sanctions - a removal of Russian banks from the SWIFT system and a withdrawal of Western banks from Russia - significantly reduce Russian trade with both Western and non-Western countries. The effects of financial sanctions are more prominent on the extensive margin, causing fewer Russian firms able to trade. Financial sanctions also amplify the effects of trade sanctions in Russian trade with Western countries. On the other hand, the effect of financial sanctions on Russian trade with non-Western countries is diminished partly by the use of non-Western currencies, particularly the Chinese Renminbi.

For further information, please contact RSFAS Seminars.

All information collected by the University is governed by the ANU Privacy Policy.

A seminar by Professor Sean Simpson from Wake Forest University

Title: TBA

Abstract: TBA

For further information, please contact RSFAS Seminars.

All information collected by the University is governed by the ANU Privacy Policy.

A seminar by Dr Jack Freestone from Macquarie University

Title: TBA

Abstract: TBA

For further information, please contact RSFAS Seminars.

All information collected by the University is governed by the ANU Privacy Policy.

A seminar by Dr Bao Anh Vu from RSFAS

Title: Estimating Antarctic subglacial conditions using neural posterior inference

Abstract: The Antarctic ice sheet (AIS) is the largest freshwater reservoir on Earth and a major contributor to sea level rise. Ice loss from the AIS will have significant impact on the biodiversity of Antarctica and its surrounding oceans as well as other coastal communities, including Australia. Therefore, estimates of the AIS's contribution to the sea-level budget are of paramount importance for planning and adaptation. Ice sheet models are routinely used to quantify and project an ice sheet's contribution to sea level rise. In order for an ice sheet model to generate realistic projections, its parameters must first be calibrated using observational data; this is challenging due to the nonlinearity of the model equations, the high dimensionality of the underlying parameters, and limited data availability for validation. This study leverages the emerging field of neural posterior approximation for efficiently calibrating ice sheet model parameters and boundary conditions. We make use of a one-dimensional (flowline) ice sheet model called the Shallow-Shelf Approximation model in a state-space framework. A neural network is trained to infer the underlying parameters, namely the bedrock elevation and basal friction coefficient along the flowline, based on observations of ice surface velocity and surface elevation. We show through a simulation study that our approach yields more accurate estimates of the parameters and states than a state-augmented ensemble Kalman filter, which is the current state-of-the-art. We then apply our approach to infer the bed elevation and basal friction along a flowline in Thwaites Glacier, Antarctica.

For further information, please contact RSFAS Seminars.

All information collected by the University is governed by the ANU Privacy Policy.

A seminar by Dr Pavel Krupskiy from The University of Melbourne

Title: Parsimonious Factor Models for Asymmetric Dependence in Multivariate Extremes

Abstract: Modelling multivariate extreme events is essential when extrapolating beyond the range of observed data. Parametric models that are suitable for real-world extremes must be flexible -- particularly in their ability to capture asymmetric dependence structures -- while also remaining parsimonious for interpretability and computationally scalable in high dimensions. Although many models have been proposed, it is rare for any single construction to satisfy all of these requirements. For instance, the popular Husler-Reiss model is limited to symmetric dependence structures. In this talk, we introduce a class of additive factor models and derive their extreme-value limits. This leads to a broad and tractable family of models characterised by a manageable number of parameters. These models naturally accommodate asymmetric tail dependence and allow for non-stationary behaviour. We present the limiting models from both the componentwise-maxima and Peaks-over-Thresholds perspectives, via the multivariate extreme value and multivariate generalized Pareto distributions, respectively. Simulation studies illustrate identifiability properties based on existing inference methodologies. Finally, applications to summer temperature maxima in Melbourne, Australia, and to weekly negative returns from four major UK banks demonstrate improved fit compared with the Husler-Reiss model.

A seminar by Dr Marina Gertsberg from University of Melbourne

Title: Editor Visits and Publication Success

Abstract: Publishing in elite journals determines tenure outcomes: a single top-five publication increases the rate of receiving tenure by 80%. Given the geographic concentration of editorial power and high stakes of publication success, understanding whether brief face-to-face meetings with editors at departmental seminars affect publication decisions matters for fairness and efficiency in academic careers. We test whether face-to-face meetings with journal editors at departmental seminars increase publication probability for assistant professors. We exploit quasi-experimental variation from seminar rotation policies, which create mechanical fluctuations in when editors visit specific schools. Using these rotation-based restrictions as an instrument for editor visits, we analyze seminar records across top finance departments linked with publication outcomes for 651 assistant professors at top 100 institutions. We find no evidence that editor visits increase publication probability. Effects are near zero across two-, three-, and four-year horizons. Our estimates have sufficient precision to rule out publication effects larger than 20% of the baseline rate at the three- and four-year horizons. The null result is robust across alternative specifications and holds for all subgroups examined (faculty gender, university rank, editor activity). The rotation instrument strongly predicts editor visits, confirming our null finding reflects genuine absence of effects rather than statistical noise. These findings establish a boundary condition for network effects in academic publishing: professional connections may help career outcomes through sustained relationships, but not through brief encounters.

For further information, please contact RSFAS Seminars.

All information collected by the University is governed by the ANU Privacy Policy

A seminar by Dr Edward Shore from UNSW

Title: ‘Did You Catch the Game Last Night?’ Peer Group Effects in Sell-Side Analyst Forecasts

Abstract: Identifying interpersonal spillovers is challenging because most shocks either affect entire groups simultaneously or may transmit information alongside behavior. We study spillovers in sell-side forecasting using a setting that isolates exposure at the individual level. Linking equity analysts to their undergraduate alma maters, we use NCAA football championship outcomes as plausibly exogenous mood shocks that affect some analysts but not their coworkers within the same brokerage. We find that 'non-winning' analysts become more optimistic when exposed to affected colleagues, consistent with sentiment spillovers rather than information transmission. These spillovers are weaker in brokerages with greater female representation. Our findings show how individually targeted, non-fundamental shocks can be used to identify spillovers inside firms.

For further information, please contact RSFAS Seminars.

All information collected by the University is governed by the ANU Privacy Policy.

A seminar by Professor Raffaele Argiento from Università degli Studi di Bergamo

Title:  Model-Based Clustering: A Bayesian Nonparametric Perspective

Abstract: Mixture models are the prototypical tool for clustering. In the Bayesian framework, a traditional distinction has been made between mixtures with a finite number of components and those with an infinite number. Recently, however, the literature has increasingly recognized that both types of mixture models can be addressed using a unified set of tools -- indeed, the clustering problem is inherently nonparametric in nature.
Adopting this perspective, in this talk, I will first present a general definition of a mixture model with a random number of components, framing it within a nonparametric context and highlighting how its probabilistic structure is naturally linked to point process theory.

Building on this connection, I will show how point process theory can be leveraged to construct repulsive or attractive mixtures. I will illustrate these ideas with an example in which the mixture model is designed to cluster categorical data, emphasizing how dependence between atoms (i.e., attractive interactions) can be exploited to study dependence structures in categorical datasets. Finally, I will discuss an application to animal taxonomy in the context of biodiversity studies. 

For further information, please contact RSFAS Seminars.

All information collected by the University is governed by the ANU Privacy Policy.