Past Seminars

Past seminars for the Academic Year 2024/25:

04 October 2024 at 16:00:

Speaker: Joshua Chan (Purdue University)

Title: Large Bayesian Matrix Autoregressions (jointly with Yaling Qi (Purdue University))

Abstract: High-dimensional matrix-valued time series are increasingly common in economics and finance. Prominent examples include large cross-region panels and dynamic economic networks. As the dimensions of the matrix grow, conventional approaches based on vector autoregressions--implemented by vectoring the matrix-valued data--become computationally infeasible. We introduce a class of large Bayesian matrix autoregressions (BMARs) that can accommodate time-varying volatility, non-Gaussian errors and COVID-19 outliers. To tackle parameter proliferation, we propose Minnesota-type shrinkage priors on the MAR coefficients. We develop a unified approach for estimating this class of models, which scales well to high dimensions. The empirical relevance of these new BMARs is illustrated using a US state-level dataset that contains 6 macroeconomic times-series for each of the 50 states, with a total of 300 times-series.

Location: Bocconi University, Via Roentgen 1, Room 4-e4-sr03 (fourth floor) 

18 October 2024 at 16:00:

Speaker: Geert Mesters (Universitat Pompeu Fabra)

Title: Innovations meet Narratives - improving the power-credibility trade-off in macro- (jointly with Regis Barnichon (San Francisco Fed))

Abstract: In empirical macro, instances of clear and indisputable exogenous variation are rare, and researchers often face a difficult trade-off between credibility and efficiency. In this work, we introduce a new method --innovation-powered IV--, which allows to reduce the confidence intervals of a credible but low power identification scheme (e.g., a narrative instrument) by leveraging the high-power of a possibly misspecified parametric identification assumption (e.g., a short run restriction). The method delivers large reductions in confidence intervals for the causal effects of monetary policy, taxes, government spending and oil price shocks, with improvements of 40 percent or more compared to state of the art narratively-identified estimates. 

Location: Bocconi University, Via Roentgen 1, ALESINA SEMINAR Room 5-e4-sr04 (fifth floor)

07 November 2024 at 12:15:

Speaker: Efrem Castelnuovo (University of Padova)

Title: Set Identification of Monetary Policy Shocks: A Trip to Monte Carlo" (jointly with Giovanni Pellegrino (University of Padova) and Laust Særkjær(Aarhus University))

Abstract: We conduct Monte Carlo simulations to assess the relative ability of sign, narrative, and policy coefficient restrictions to identify the output effects of monetary policy shocks. We find narrative restrictions to substantially improve the performance of the sign-retriction-only approach. We show that policy coefficient restrictions can work in favor of sharpening the identification of the output effects of monetary policy shocks based on sign and narrative restrictions only. When applied to Uhlig's (2005) reference dataset, a combination of sign, narrative, and policy coefficient restrictions calls for a fairly precisely estimated drop in output of about -5% after one year. This drop is twice as large as the one obtained with sign and narrative restrictions only.

Location: University of Milan  -Seminar Room (2nd Floor), Via Conservatorio 7

19 November 2024 at 12:00:

Speaker: Esther Ruiz Ortega (Universidad Carlos III de Madrid)

Title: Dealing with idiosyncratic cross-correlation when constructing confidence regions for PC factors: Relevance for constructing inflation scenarios (jointly with Diego Fresoli and Pilar Poncela (Univ. Autónoma de Madrid )

Abstract: In this paper, we propose a computationally simple estimator of the asymptotic covariance matrix of the Principal Components (PC) factors valid in the presence of cross-correlated idiosyncratic components. The proposed estimator of the asymptotic Mean Square Error (MSE) of PC factors is based on adaptive thresholding the sample covariances of the idiosyncratic residuals with the threshold based on their individual variances. We compare the finite sample performance of confidence regions for the PC factors obtained using the proposed asymptotic MSE with those of available extant asymptotic and bootstrap regions and show that the former beats all alternative procedures for a wide variety of idiosyncratic cross-correlation structures. 

Location: FEEM, Corso Magenta 63, Sala Cinema

05 December 2024 at 12:15:

Speaker: Michele Piffer (King's College London and Bank of England)

Title: Non-Gaussian Business Cycle Analysis (joint with Christian Matthes (Indiana University) and Andrzej Kocięcki (University of Warsaw)

Abstract: Which shocks explain the volatility in US real GDP? We study this question using a vector autoregressive model with t-distributed shocks. We first show that a simple reparameterization allows for the development of the first Gibbs sampler for this model. This improves upon existing methods that require a computationally more demanding Metropolis-Hastings step, allowing us to use larger VAR models than in the previous literature. Our application to US data suggests that there is no such thing as a single, main business cycle shock. No shock explains more than 20% of the variability of real GDP, with the largest role played by a weakly inflationary demand shock.

Location: University of Milan  -Seminar Room (2nd Floor), Via Conservatorio 7

11 December 2024 at 12:00:

Speaker: Raffaella Giacomini (University College London)

Title: Perceived shocks and impulse responses

Abstract: This paper shows how the information present in many expectations datasets can be leveraged in a new way in order to extract empirical measures of beliefs about key economic quantities: shocks and their dynamic effects. The information needed is a panel of expectations revisions of one variable across different horizons and over time. The idea is to fit a (time-varying) factor model to the revisions and obtain nonparametric estimates of the latent shocks (the factors) and the associated impulse responses (the loadings). The method relies on weak assumptions and deals with the small-sample nature of these data. An application to consensus inflation expectations reveals a single perceived shock that is highly correlated with inflation surprises and time-varying impulse responses that imply a secular decrease in the perceived persistence of the shock (that is, more “anchored” long-term inflation expectations).

Location: University of Milano-Bicocca, Second floor of the U7 building, room 2104.

16 January 2025 at 12:00:

Speaker: Matteo Barigozzi (University of Bologna)

Title: Predicting Energy Demand with Matrix and Tensor Factor Models

Abstract: This work provides a novel framework for modeling time series displaying multiple seasonal patterns. The methodology presented builds on recent advancements in high-dimensional factor analysis, focusing on time series tensor factor models. The method is applied in a the domain of energy demand forecasting, considering hourly data of energy demand in the U.S. We show that the proposed method effectively captures the multi-seasonal patterns in the data, providing interpretable loading values in line with the expected characteristics of the underlying phenomena. Albeit the extraction of seasonal components is achievable through the simpler matrix factor models, we argue that a tensor factor model provide stronger asymptotic properties based on a thoughtful extension of the original dataset encompassing multiple electricity providers. Tensor factor models’ results are evaluated against classical vector factor models and functional time series methods, demonstrating the superior forecasting accuracy of the tensor approach. The analyses in this work provide a robust framework for future model extensions by effectively accounting for complex seasonal patterns. These models can be integrated into more complex empirical settings, allowing for the incorporation of additional variables to enhance accuracy in diverse forecasting scenarios.

Location: FEEM, Corso Magenta 63, Sala Cinema

06 February 2025 at 12:15:

Speaker: Toru Kitagawa (Brown University)

Title: Policy Choice in Time Series by Empirical Welfare Maximization

Abstract: This paper develops a novel method for policy choice in a dynamic setting where the available data is a multivariate time series. Building on the statistical treatment choice framework, we propose Time-series Empirical Welfare Maximization (T-EWM) methods to estimate an optimal policy rule by maximizing an empirical welfare criterion constructed using nonparametric potential outcome time series. We characterize conditions under which T-EWM consistently learns a policy choice that is optimal in terms of conditional welfare given the time-series history. We derive a nonasymptotic upper bound for conditional welfare regret. To illustrate the implementation and uses of T-EWM, we perform simulation studies and apply the method to estimate optimal restriction rules against Covid-19.

Location: University of Milan  -Seminar Room (2nd Floor), Via Conservatorio 7

21 February 2025 at 12:15 - Cancelled! 

Speaker: Evi Pappa (Universidad Carlos III de Madrid)

Location: Politecnico of Milan (Bovisa) - Via R. Lambruschini 4/B 

17 March 2025 at 16:00:

Speaker: Simone Manganelli (European Central Bank)

Title: Improving benchmark asset allocations with judgment

Abstract: Integrating the concept of judgment into a statistical decision framework addresses longstanding empirical challenges in asset allocation models. Judgment is defined by a benchmark allocation and a level of significance. The null hypothesis posits that the benchmark allocation is optimal if its gradient of the expected utility is zero. A test statistic evaluates the gradient using the maximum likelihood estimate. If the test does not reject the null hypothesis at the specified significance level, the benchmark allocation is retained. Otherwise, decision-makers select the allocation identified by the boundary of the confidence interval of the empirical gradient. Monte Carlo simulations and pseudo out-of-sample exercises indicate that this judgment-based approach outperforms traditional methods and the 1/N benchmark, providing a robust alternative for asset allocation under uncertainty.

Location: Bocconi University

26 March 2025 at 10:00:

Speaker: Giovanni Urga (Bayes Business School, City University of London)

Title: Macroeconomic Announcements, Confidence and Economic Activity Through the Business Cycles

Abstract: We propose two new indices — the Business Confidence Surprise Index and the Industrial Production Surprise Index — to explore the relationship between economic confidence and business cycles across seven major economies (United States, Japan, United Kingdom, Germany, France, Italy, Spain) from 2000 to 2023. Using a Vector Autoregressive framework, we examine how macroeconomic surprise shocks propagate through economic confidence and activity. We find that macroeconomic surprise shocks have a lasting influence on economic activity. The long-term effects are significantly stronger during periods of financial distress and low economic confidence, suggesting heightened sensitivity to such shocks in adverse conditions.

Location: University of Milano-Bicocca, Second floor of the U7 building, room 2104.

31 March 2025 from 9:00 to 18:00:

Junior Milan Time Series Workshop (Program)

Keynote speaker: Roberto Casarin (Ca' Foscari University of Venice)

Location: Sala Lauree - University of Milan, Via Conservatorio 7, 20122, Milan

17 June 2025 at 12:15:

Speaker: Sophocles Mavroeidis (University of Oxford)

Title: TBA

Abstract: TBA

Location: Politecnico di Milano - Room BL27.14 (Bovisa - La Masa - Via Lambruschini 4)

Past seminars for the Academic Year 2023/24:

12 October 2023: 

Speaker: Alessandra Luati  (University of Bologna and Imperial College London)

Title: On the optimality of score-driven models (jointly with P. Gorgi and S. Lauria) 

Abstract: Score-driven models have been recently introduced as a general framework to specify time-varying parameters of conditional densities. The score enjoys stochastic properties that make these models  easy to implement and convenient to apply in several contexts, ranging from biostatistics to finance. Score-driven parameter  updates have been shown to be  optimal in terms of locally reducing  a local version of the Kullback–Leibler divergence between the true  conditional density and the  postulated density of the model.  A key limitation of such optimality property is that it holds only locally both in the parameter space and sample space, yielding to a definition of  local  Kullback–Leibler divergence that is  in fact  not a divergence measure. The current paper shows that score-driven   updates satisfy  stronger optimality properties that are based on a global definition of  Kullback–Leibler divergence. In particular, it is shown  that score-driven updates reduce the  distance between the expected updated parameter and the pseudo-true parameter. Furthermore, depending on the conditional density and the scaling of the score, the optimality result can hold globally over the parameter space, which can be viewed as a generalisation of the monotonicity property of the stochastic gradient descent scheme. Several examples illustrate how the results derived in the paper apply to specific models under different  easy-to-check assumptions, and  provide a formal method to select the link-function  and  the scaling  of the score.

Location: University of Milan.

24 October 2023 at 16:30:

Speaker: Daniel J. Lewis (University College London (UCL))

Title: A Robust Test for Weak Instruments with Multiple Endogenous Regressors (with Karel Mertens) 

Abstract: We extend the popular bias-based test of Stock and Yogo (2005) for instrument strength in linear instrumental variables regressions with multiple endogenous regressors to be robust to heteroskedasticity and autocorrelation. Equivalently, we extend the robust test of Montiel Olea and Pflueger (2013) for one endogenous regressor to the general case with multiple endogenous regressors. We describe a simple procedure for applied researchers to conduct our generalized first-stage test of instrument strength and provide efficient and easy-to-use Matlab code for its implementation. We demonstrate our testing procedures by considering the estimation of the state-dependent effects of fiscal policy as in Ramey and Zubairy (2018).

Location: Bocconi University.

6 November 2023 at 16:30:

Speaker: Florian Huber (University of Salzburg)

Title: Bayesian Nonlinear Regression using Sums of  Simple Functions

Location: Bocconi University - Room 5.e4.sr04 (Alesina Seminar Room). Floor 5, Via Roentgen, 1, Milan

13 November 2023 at 14:30:

Speaker: Fabio Canova (BI Norwegian Business School)

Title: What Drives the Recent Surge in Inflation? The historical Decomposition Roller Coaster (with D. Bergholt, F. Furlanetto, N. Maffei-Faccioli and P. Ulvedal)

Abstract: What drives the inflation surge in the post Covid period? To answer this question, one must decompose observable fluctuations into the contributions due to structural shocks. We document how whimsical an historical shock decomposition can be in standard vector autoregressive models. Neglecting the uncertainty surrounding the deterministic component of the model implies implausible behavior for shocks over history under general conditions. Our favorite approach to solve the problem, the single unit prior, shrinks the massive uncertainty around the deterministic components toward their sample mean values. With such a prior and a standard sign-identified VAR, demand shocks are the main drivers of the current inflation surge, both in the United States and in the euro area.

Location: University of Milan-Bicocca.

13 December 2023 at 12:15:

Speaker: Lorenzo Trapani (University of Leicester and University of Pavia)

Title: Changepoint detection in functional data with empirical energy distance (joint with BC Boniece and L Horvath)

Location: University of Milan-Bicocca - Room 2104 (2nd Floor), Building U7, Piazza Ateneo Nuovo, Milan

08 February 2024 at 12:15:

Speaker: Andreas Pick (Erasmus University of Rotterdam and De Nederlandsche Bank)

Title: Forecasting with panel data: Estimation uncertainty versus parameter heterogeneity (with M.H. Pesaran and A. Timmermann)

Abstract: We provide a comprehensive examination of the predictive accuracy of panel forecasting methods based on pooling, random effects, fixed effects and Bayesian estimation relative to forecasts based on individual estimates. We also propose optimal weights for forecast combination schemes. We demonstrate how predictive accuracy depends on the degree of parameter heterogeneity, whether such heterogeneity is correlated with the regressors, the goodness of fit of the model, and the cross-sectional (N) and time (T) dimensions. Forecasting performance is examined through Monte Carlo simulations and empirical applications to house prices and CPI inflation. We find that forecast combination and Bayesian forecasting methods perform best overall and rarely produce the least accurate forecasts for individual series.

Location: University of Milan  -Seminar Room (2nd Floor), Via Conservatorio 7

21 March 2024 at 12:15:

Speaker: Marta Banbura (European Central Bank)

Title: Advances in Modeling Time-Varying Trends using Large VARs: Order-Invariant Stochastic Volatility, Hierarchical Shrinkage and Outliers

Abstract: Measuring macroeconomic trends in a rapidly changing environment is challenging, as it is difficult to disentangle abrupt changes in trends from outliers. This paper tackles this challenge by developing a novel steady-state Bayesian VAR with a number of important features. First, the model incorporates an hierarchical shrinkage prior on the time-varying trends that favors smooth trend transitions, while it is also capable of detecting abrupt changes. Second, it features an outlier component that can address extreme observations such as COVID-19 outliers. Third, it builds upon an order-invariant stochastic volatility specification, as opposed to the commonly-used Cholesky-based stochastic volatility models under which trend estimates may depend on how the endogenous variables enter the system. We illustrate the methodology using US and EA disaggregated inflation data.

Location: University of Milan  -Seminar Room (2nd Floor), Via Conservatorio 7

11 April 2024 at 12:15:

Speaker: Anna Simoni (CNRS, French National Center of Scientific Research and CREST)

Title: Bayesian Bi-level Sparse Group Regressions for Macroeconomic Forecasting (joint with Matteo Mogliani)

Abstract: We propose a Machine Learning approach for optimal macroeconomic forecasting in a high-dimensional setting with covariates presenting a known group structure. Our model encompasses forecasting settings with many series, mixed frequencies, and unknown nonlinearities. We introduce in time-series econometrics the concept of bi-level sparsity, i.e. sparsity holds at both the group level and within groups, and we assume the true model satisfies this assumption. We propose a prior that induces bi-level sparsity, and the corresponding posterior distribution is demonstrated to contract at the minimax-optimal rate, recover the model parameters, and have a support that includes the support of the model asymptotically. Our theory allows for correlation between groups, while predictors in the same group can be characterized by strong covariation as well as common characteristics and patterns. Finite sample performance is illustrated through comprehensive Monte Carlo experiments and a real-data nowcasting exercise of the US GDP growth rate. 

Location: University of Milan  -Seminar Room (2nd Floor), Via Conservatorio 7

15 May 2024 at 12:00:

Speaker: Matteo Ciccarelli (European Central Bank)

Title: The macroeconomic effects of climate change and the green transition

Abstract: Modelling interaction between climate change, the green transition and the macroeconomy requires empirically validated assumptions. The presentation will focus on a few questions of relevance from the perspective of central banks and on the importance to understand the business cycle effects of climate-related shocks and risks. Based on recent empirical and modelling work on the topic and through the lens of standard economic concepts and models, the presentation will shed light on the demand and supply channels of climate-related shocks, on the importance of appropriately choosing the variables and proxies to ‘identify’ the shocks and account for the transmission channels in empirical analysis, and on the asymmetric and heterogeneous effects of weather shocks and climate policies across countries. The presentation will conclude with implications for modellers and policy makers.

Location: FEEM, Corso Magenta 63

26 June 2024 at 12:15:

Speaker: Filippo Ferroni (Federal Reserve Bank of Chicago)

Title: Higher-Order Moment Inequality Restrictions for SVARs 

Abstract: We exploit inequality restrictions on higher-order moments of the distribution of structural shocks to sharpen their identification. We show that these constraints can be treated as necessary conditions and used to shrink the set of admissible rotations. We illustrate the usefulness of this approach showing, by simulations, how it can dramatically improve the identification of monetary policy shocks when combined with widely used sign-restriction schemes. We then apply our methodology to two empirical questions: the effects of monetary policy shocks in the U.S. and the effects of sovereign bond spread shocks in the euro area. In both cases, using higher-moment restrictions significantly sharpens identification. After a shock to euro area government bond spreads, monetary policy quickly turns expansionary, corporate borrowing conditions worsen on impact, the real economy and the labor market of the euro area contract appreciably, and returns on German government bonds fall, likely reflecting investors’ flight to quality. 

Location: Politecnico of Milan (Bovisa) - Via R. Lambruschini 4/B