Econometrics Seminars
Below are the econometrics seminars scheduled at the School of Economics, University of Sydney. If you would like to find more information or would like to present in our seminars, please contact our seminar coordinator Dakyung Seong (dakyung.seong@sydney.edu.au)
2023 Semester 2 (Aug-Nov, 2023)
All the seminars this semester are back to in-person only, and will be held in Social Sciences Building (SSB), University of Sydney.
August 2: Dan Millimet, Southern Methodist University. (SSB 650, 2-3:30pm)
Title: Fixed Effects and Causal Inference
Abstract: With data spanning multiple units and time periods, the fixed effects estimator is the default estimator of causal effects with linear models and nonexperimental data due to the removal of time-invariant heterogeneity. One feature of the estimator is often overlooked, however: the attributes that are time-invariant are not invariant to the length of the panel. We consider several alternatives to the fixed effects estimator with more than two time periods when relevant unit-specific heterogeneity is not time-invariant, including novel rolling estimators. Through simulations and replications, we find little gain —and much to lose— by using the fixed effects estimator.
August 16: Ayden Higgins, University of Oxford. (SSB 650, 2-3:30pm)
Title: Instrumental Variables for Dynamic Spatial Models with Interactive Fixed Effects
Abstract: This paper studies estimation and inference in the context of a dynamic spatial model with interactive fixed effects. A simple instrumental variables interactive fixed effects (IV-IFE) estimator is proposed which can be used to produce consistent and asymptotically unbiased estimates with T fixed and n tends to infinity. However, in practice, this estimator can suffer from significant finite sample bias. Further investigation reveals that the source of this problem lies with a particular term in the asymptotic expansion which may diminish only very slowly as sample size increases. To remedy this, a bias corrected estimator is constructed, which, through extensive simulation, is demonstrated to exhibit significantly improved finite sample performance.
August 23: Wei Huang, University of Melbourne. (SSB 441, 2-3:30pm)
Title: Nonparametric Estimation of the Continuous Treatment Effect with Measurement Error
Paper link: https://doi.org/10.1093/jrsssb/qkad013
September 6: Dick Startz, University of California Santa Barbara. (SSB 650, 2-3:30pm)
Title: Recessions, Recoveries, and Leverage
Abstract: When leverage is low, recoveries from recessions are likely to eventually return the economy to its pre-recession growth path. When leverage is high, recoveries are likely to leave the economy below its pre-recession growth path. In other words, low-leverage recessions are likely to be U-shaped while high-leverage recessions are likely to be L-shaped. The increase in leverage over the post-War period that recent recessions are much more likely to be L-shaped. In particular, there is strong evidence that the Great Recession was L-shaped. We find similar effects of leverage for a number of other countries, but not all.
October 4: Zhenting Sun, Peking University. (SSB 650, 2-3:30pm) -- Cancelled
October 12: Dacheng Xiu, University of Chicago. (SSB 650, 11am-12:30pm)
Title: Prediction When Factors are Weak
Abstract: In macroeconomic forecasting, principal component analysis (PCA) has been the most prevalent approach to the recovery of factors, which summarize information in a large set of macro predictors. Nevertheless, the theoretical justification of this approach often relies on a convenient and critical assumption that factors are pervasive. To incorporate information from weaker factors, we propose a new prediction procedure based on supervised PCA, which iterates over selection, PCA, and projection. The selection step finds a subset of predictors most correlated with the prediction target, whereas the projection step permits multiple weak factors of distinct strength. We justify our procedure in an asymptotic scheme where both the sample size and the cross-sectional dimension increase at potentially different rates. Our empirical analysis highlights the role of weak factors in predicting inflation, industrial production growth, and changes in unemployment.
Paper link: https://papers.ssrn.com/sol3/papers.cfm?abstract_id=4397485
October 18: Ho Leung Ip, Charles Sturt University. (SSB 650, 2-3pm)
Title: A Mixture Distribution for Modelling Bivariate Ordinal Data
Abstract: Ordinal responses often arise from surveys which require respondents to rate items on a Likert scale. Since most surveys contain more than one question, the data collected are multivariate in nature, and the associations between different survey items are usually of considerable interest. In this presentation, a mixture distribution, called the combination of uniform and binomial (CUB), will be revisited. Under CUB, each response is assumed to originate from either the respondent's uncertainty or the actual feeling towards the survey item. An extension of the univariate CUB model to the bivariate case will be introduced. The proposed model allows the associations between the unobserved uncertainty and feeling components of the responses to be estimated, a distinctive feature compared to previous attempts using copula-based approaches. This presentation will describe the underlying logic and both theoretical and practical aspects of the proposed model.
November 1: Xiaotian Zheng, University of Wollongong. (SSB 650, 2-3pm)
Title: Mixture Modelling for Temporal Point Processes with Memory
Abstract: In this talk, I will present a constructive approach to building temporal point processes that incorporate dependence on their history. The dependence is modelled through the conditional density of the duration, i.e., the interval between successive event times, using a mixture of first-order conditional densities for each one of a specific number of lagged durations. Such a formulation for the conditional duration density accommodates high-order dynamics, and the implied conditional intensity function admits a representation as a local mixture of first-order hazard functions. By specifying appropriate families of distributions for the first-order conditional densities, with different shapes for the associated hazard functions, we can obtain either self-exciting or self-regulating point processes. From the perspective of duration processes, we develop a method to specify a stationary marginal density. The resulting model, interpreted as a dependent renewal process, introduces high-order Markov dependence among identically distributed durations. Furthermore, we provide extensions to cluster point processes. These can describe duration clustering behaviors attributed to different factors, thus expanding the scope of the modelling framework to a wider range of applications. The proposed methodology is illustrated with data examples from environmental science and finance.
November 22: Yongmiao Hong, University of Chinese Academy of Sciences. (SSB 650, 2-3:30pm)
Title: Time-Varying Factor Selection: A Sparse Fused GMM Approach
Abstract: Empirical asset pricing studies evaluate and select risk factors solely based on their historical aggregate performance, implicitly assuming a time-invariant model specification, and overlooking potential time variations of specification in the stochastic discount factor (SDF) model. This paper presents a new method for capturing the time-varying sparsity of factor models by identifying heterogeneous structural breaks instrumented by macroeconomic regimes. Our empirical findings highlight that factor model specification changes over time. We identify time-invariant factors as well as time-varying factors, selected in different periods in response to macroeconomic-targeted regime switching. The collective explanatory power of these 20 risk factors is high during periods of high interest rates or low market valuation, but their effectiveness declines when market liquidity is high. Finally, we evaluate factors by modeling unsynchronized factor discovery using unbalanced panel data to account for heterogeneous academic publication timings.
November 29: Massimo Franchi, Sapienza University of Rome.
Title: Estimation and inference on stochastic trends in large dimensional systems
Abstract: The talk will review the recent literature on stochastic trends and cointegration in large dimensional multivariate (possibly functional) time series and present new results on estimation and inference in these systems. The limit distributions of empirical eigenvalues and eigenvectors associated with alternative novel Canonical Correlations Analyses (CCA) will be discussed. It will be shown how these CCA deliver estimators of the number of common trends and of a basis of the common trends loadings space. (Super-)consistency as well as the asymptotic distributions of estimators are derived. The properties of the estimators are compared with existing alternatives both theoretically and via Monte Carlo simulations.
Past seminars
2023 Semester 1
All the seminars this semester are held in-person with Zoom link available.
March 8: Bonsoo Koo, Monash University. "What Impulse Response Do Instrumental Variables Identify?".
March 22: James Duffy, University of Oxford. "Cointegration with Occasionally Binding Constraints".
April 5: Hugo Freeman, University College London. "Multidimensional Interactive Fixed-Effects".
April 26: Jiti Gao, Monash University. "A Unified Approach to Estimating Time-Varying Trends".
May 10: Hanlin Shang, Macquarie University. "Detecting structural breaks in high-dimensional functional time series".
May 24: Yanrong Yang, Australian National University. "Eigen-analysis of high-dimensional time series".
May 31: Fu Ouyang, Queensland University. "High-Dimensional Binary Choice Models with Unknown Heteroskedasticity or Instrumental Variables".
2022 Semester 2
All the seminars this semester are held in-person with Zoom link available. In the following, "SSB" stands for Social Sciences Building (A02), University of Sydney.
August 2: Colin Cameron, University of California Davis. "Recent Developments in Cluster-Robust Inference".
August 10: Adam Clements, Queensland University of Technology. "Using Threshold Style Volatility Measures for Estimating HAR Model Coefficients".
August 24: Nan Zou, Macquarie University. "Bootstrap Massive/Chaotic Data".
August 25: Tim Christensen, University College London and New York University. "Adaptive Estimation and Uniform Confidence Bands for Nonparametric Structural Functions and Elasticities".
September 7: Firmin Doko Tchatoka, University of Adelaide. "Relevant Moment Selection Under Mixed Identification Strength".
September 21: Andrew Patton, Duke University. "Testing Forecast Rationality of Measures of Central Tendency".
October 12: Won-Ki Seo, University of Sydney. "Inference on nonstationarity and common trends in high-dimensional or functional time series".
October 27: James Duffy, University of Oxford. "The Cointegrated VAR without Unit Roots".
November 2: Tang Srisuma, National University of Singapore. "Identification and Estimation of a Search Model: A Procurement Auction Approach" .
November 16: Thomas Tao Yang, Australian National University.
2022 Semester 1
March 23: Wooyong Lee, University of Technology Sydney. "Identification and Estimation of Dynamic Random Coefficient Models" [link].
March 30: Feng Chen, University of New South Wales. "Renewal Hawkes Processes" [link].
April 6: Zhenting Sun, Peking University. "Pairwise Valid Instruments" [link].
April 20: Aurore Delaigle, University of Melbourne. "Estimating a Covariance Function from Fragments of Functional Data" [link].
April 27: Dakyung Seong, University of Sydney. "Functional Instrumental Variable Regression with an Application to Estimating the Impact of Immigration on Native Wages" [link].
May 11: Simon Kwok, University of Sydney. "A Consistent and Robust Test for Autocorrelated Jump Occurrences" [link].
May 25: Yichong Zhang, Singapore Management University. "Regression-Adjusted Estimation of Quantile Treatment Effects under Covariate-Adaptive Randomizations" [link].
2019 Semester 2
August 5: Colin Cameron, UC Davis. "Machine learning in economics".
August 29: Michael Fan, Xiamen University. "Estimation of conditional average treatment effects with high-dimensional data".
September 5: Morten Ø. Nielsen, Queen’s University. "Wild bootstrap and asymptotic inference with multiway clustering" [link].
September 12: Dick van Dijk, Erasmus University Rotterdam. "Uncertainty and the macroeconomy: A real-time out-of-sample evaluation" [link].
September 16: Arthur Lewbel, Boston College. "Social networks with misclassified or unobserved links" [link].
October 21: Shuping Shi, Macquarie University. "Common bubble detection in large dimensional financial systems" [link].
November 4: Didier Nibbering, Monash University. "Panel forecasting with asymmetric grouping" [link].
November 14: Massimo Franchi, Sapienza Università di Roma. "Autoregressive processes, cointegration and related results" [link].
November 18: David Frazier, Monash University. "Weak instruments in discrete choice models" [link].
Other seminars down the street and across town: