Research Interests:
Macroeconomic modeling, measurement, and forecasting • Uncertainty, expectations, and adaptive learning • Time series econometrics • Complex systems and finance
Links:
RePEc/IDEAS • CAMA/ANU • KOF/ETH-Z • ORCiD • LinkedIn • Google Scholar
Galimberti, J. K. (2024) Initial Beliefs Uncertainty. B.E. Journal of Macroeconomics, 24(1), pp. 45-96.
Abstract: This paper evaluates how initial beliefs uncertainty can affect data weighting and the estimation of models with adaptive learning. One key finding is that misspecification of initial beliefs uncertainty, particularly with the common approach of artificially inflating initials uncertainty to accelerate convergence of estimates, generates time-varying profiles of weights given to past observations in what should otherwise follow a fixed profile of decaying weights. The effect of this misspecification, denoted as diffuse initials, is shown to distort the estimation and interpretation of learning in finite samples. Simulations of a forward-looking Phillips curve model indicate that (i) diffuse initials lead to downward biased estimates of expectations relevance in the determination of actual inflation, and (ii) these biases spill over to estimates of inflation responsiveness to output gaps. An empirical application with U.S. data shows the relevance of these effects for the determination of expectational stability over decadal subsamples of data. The use of diffuse initials is also found to lead to downward biased estimates of learning gains, both estimated from an aggregate representative model and estimated to match individual expectations from survey expectations data.
Keywords: expectations, adaptive learning, bounded rationality, macroeconomics. JEL codes: E70, D83, D84, E37, C32, C63.
Links: Working paper
Galimberti, J.K. & Pichler, S. & Pleninger, R. (2023) Measuring Inequality using Geospatial Data, World Bank Economic Review, 37(4), pp. 549-569.
Abstract: The main challenge in studying inequality is limited data availability, which is particularly problematic in developing countries. This study constructs a measure of light-based geospatial income inequality (LGII) for 234 countries/territories from 1992 to 2013 using satellite data on night lights and gridded population data. Key methodological innovations include the use of varying levels of data aggregation, and a calibration of the lights-prosperity relationship to match traditional inequality measures based on income data. The new LGII measure is significantly correlated with cross-country variation in income inequality. Within countries the light-based inequality measure is also correlated with measures of energy efficiency and the quality of population data. Two applications of the data are provided in the fields of health economics and international finance. The results show that light- and income-based inequality measures lead to similar results, but the geospatial data offers a significant expansion of the number of observations.
Keywords: nighttime lights, inequality, gridded population. JEL codes: D63, E01, I14, O11, O47, O57.
Galimberti, J.K. (2020) Forecasting GDP growth from outer space, Oxford Bulletin of Economics and Statistics, 82(4), pp. 697-722.
Abstract: We evaluate the usefulness of satellite-based data on night-time lights for forecasting GDP growth across a global sample of countries, proposing innovative location-based indicators to extract new predictive information from the lights data. Our findings are generally favorable to the use of night lights data to improve the accuracy of model-based forecasts. We also find a substantial degree of heterogeneity across countries in the relationship between lights and economic activity: individually-estimated models tend to outperform panel specifications. Key factors underlying the night lights performance include the country's size and income level, logistics infrastructure, and the quality of national statistics.
Keywords: night lights, remote sensing, big data, business cycles, leading indicators. JEL codes: C55, C82, E01, E37, R12.
Links: latest WP version.
Media coverage: KOF Bulletin, Bilanz, Business Insider France.
Galimberti, J.K. (2019) An approximation of the distribution of learning estimates in macroeconomic models, Journal of Economic Dynamics & Control, 102, p. 29-43.
Abstract: Adaptive learning under constant-gain allows persistent deviations of beliefs from equilibrium so as to more realistically reflect agents' attempt of tracking the continuous evolution of the economy. A characterization of these beliefs is therefore paramount to a proper understanding of the role of expectations in the determination of macroeconomic outcomes. In this paper we propose a simple approximation of the first two moments (mean and variance) of the asymptotic distribution of learning estimates for a general class of dynamic macroeconomic models under constant-gain learning. Our approximation provides renewed convergence conditions that depend on the learning gain and the model's structural parameters. We validate the accuracy of our approximation with numerical simulations of a Cobweb model, a standard New-Keynesian model, and a model including a lagged endogenous variable. The relevance of our results is further evidenced by an analysis of learning stability and the effects of alternative specifications of interest rate policy rules on the distribution of agents' beliefs.
Keywords: expectations, adaptive learning, constant-gain, policy stability. JEL codes: D84, E03, E37, C62, C63.
Links: Working paper.
Berardi, M. & Galimberti, J.K. (2019) Smoothing-based initials for learning-to-forecast algorithms, Macroeconomic Dynamics, 23(3), p. 1008-1023.
Abstract: Under adaptive learning, recursive algorithms are proposed to represent how agents update their beliefs over time. For applied purposes these algorithms require initial estimates of agents perceived law of motion. Obtaining appropriate initial estimates can become prohibitive within the usual data availability restrictions of macroeconomics. To circumvent this issue we propose a new smoothing-based initialization routine that optimizes the use of a training sample of data to obtain initials consistent with the statistical properties of the learning algorithm. Our method is generically formulated to cover different specifications of the learning mechanism, such as the Least Squares and the Stochastic Gradient algorithms. Using simulations we show that our method is able to speed up the convergence of initial estimates in exchange for a higher computational cost.
Keywords: learning algorithms, initialization, smoothing, expectations. JEL codes: C63, D84, E37.
Links: pre-print.
Berardi, M. & Galimberti, J.K. (2017) Empirical calibration of adaptive learning, Journal of Economic Behavior & Organization, 144, p. 219-237.
Abstract: Adaptive learning introduces persistence in the evolution of agents' beliefs over time, helping explain why economies present sluggish adjustments towards equilibrium. The pace of this learning process is directly determined by the gain parameter. We document and evaluate gain calibrations for a broad range of model specifications with macroeconomic data, also developing alternative approaches to the endogenous determination of time-varying gains in real-time. Our key findings are that learning gains are higher for inflation than for output growth and interest rates, and that calibrations to match survey forecasts are lower than those derived according to forecasting performance, suggesting some degree of bounded rationality in the speed with which agents update their beliefs.
Keywords: bounded rationality, expectations, forecasting, real-time data, recursive estimation. JEL codes: D83, E03, E37.
Links: WP.
Berardi, M. & Galimberti, J.K. (2017) On the initialization of adaptive learning in macroeconomic models, Journal of Economic Dynamics & Control, 78, p. 26-53.
Abstract: We review and evaluate methods previously adopted in the applied literature of adaptive learning in order to initialize agents' beliefs. Previous methods are classified into three broad classes: equilibrium-related, training sample-based, and estimation-based. We conduct several simulations comparing the accuracy of the initial estimates provided by these methods and how they affect the accuracy of other estimated model parameters. We find evidence against their joint estimation with standard moment conditions: as the accuracy of estimated initials tends to deteriorate with the sample size, spillover effects also deteriorate the accuracy of the estimates of the model's structural parameters. We show how this problem can be attenuated by penalizing the variance of estimation errors. Even so, the joint estimation of learning initials with other model parameters is still subject to severe distortions in small samples. We find that equilibrium-related and training sample-based initials are less prone to these issues. We also demonstrate the empirical relevance of our results by estimating a New Keynesian Phillips curve with learning, where we find that our estimation approach provides robustness to the initialization of learning. That allows us to conclude that under adaptive learning the degree of price stickiness is lower compared to inferences under rational expectations.
Keywords: expectations, adaptive learning, initialization, algorithms, hybrid New Keynesian Phillips curve. JEL codes: C63, D84, E03, E37.
Links: WP.
Galimberti, J.K. & Suhadolnik, N. & Da Silva, S. (2017) Cowboying stock market herds with robot traders, Computational Economics, 50, p. 393-423.
Abstract: One explanation for large stock market fluctuations is its tendency to herd behavior. We put forward an agent-based model where instabilities are the result of liquidity imbalances amplified by local interactions through imitation, and calibrate the model to match some key statistics of actual daily returns. We show that an "aggregate market-maker" type of liquidity injection is not successful in stabilizing prices due to the complex nature of the stock market. To offset liquidity shortages, we propose the use of locally triggered contrarian rules, and show that these mechanisms are effective in preventing extreme returns in our artificial stock market.
Keywords: herding, robot trading, financial regulation, agent-based model. JEL codes: C63, G02.
Links: pre-print.
Galimberti, J.K. & Moura, M.L. (2016) Improving the reliability of real-time output gap estimates using survey forecasts, International Journal of Forecasting, 32, p. 358-373.
Abstract: Measuring economic activity in real-time is a crucial issue both in applied research and in the decision-making process of policy makers; however, it also poses intricate challenges to statistical filtering methods that are built to operate optimally when working with an infinite number of observations. In this paper, we propose and evaluate the use of survey forecasts for augmenting such methods, in order to reduce the end-of-sample uncertainty that is observed in the resulting gap estimates.Wefocus on three filtering methods that are employed commonly in business cycle research: the Hodrick-Prescott filter, unobserved components models, and the band-pass filter. We find that the use of surveys achieves powerful improvements in the real-time reliability of the economic activity measures associated with these filters, and argue that this approach is preferable to model-based forecasts due to both its usually superior accuracy in predicting current and future states of the economy and its parsimony.
Keywords: business cycles measurement, end-of-sample uncertainty, gap and trend decomposition. JEL codes: E32, E37.
Links: pre-print.
Berardi, M. & Galimberti, J.K. (2013) A note on exact correspondences between adaptive learning algorithms and the Kalman filter, Economics Letters, 118, p. 139-142.
Abstract: We extend the correspondences between adaptive learning algorithms and the Kalman filter to formulations with time-varying gains. Our correspondences hold exactly, in a computational implementation sense, and we discuss how they relate to previous approximate correspondences found in the literature.
Keywords: adaptive learning, least squares, stochastic gradient, Kalman filter. JEL codes: C32, C63, D83, D84.
Galimberti, J.K. & Moura, M. (2013) Taylor rules and exchange rate predictability in emerging economies, Journal of International Money & Finance, 32, p. 1008-1031.
Abstract: This study demonstrates the relationship between exchange rate determination and an endogenous monetary policy represented by Taylor rules. We fill a gap in the literature by focusing on a group of fifteen emerging economies that adopted free-floating exchange rates and inflation targeting beginning in the mid-1990s. Because of the limited span of the time series, which is a common obstacle to studying emerging economies, we employ panel data regressions to produce more efficient estimates. Following the recent literature, we use a robust set of out-of-sample statistics, incorporating bootstrapped and asymptotic distributions for the Diebold-Mariano statistic, the Clark and West statistic and Theil's U ratio. By evaluating different specifications for the Taylor rule exchange rate model based on their out-of-sample performances, we find that a present-value forward-looking specification shows strong evidence of exchange rate predictability.
Keywords: Taylor rule exchange rate model, forecasting, emerging economies, panel data, bootstrap. JEL codes: F31, F37, F41, F47.
Galimberti, J.K. & Da Silva, S. (2012) An empirical case against the use of genetic-based learning classifier systems as forecasting devices, Economics Bulletin, 32, p. 354-369.
Abstract: We adapt a genetic-based learning classifier system to a forecast evaluation exercise by making its key parameters endogenous and taking into account the need of convergence of the learning algorithm, an issue usually neglected in the literature. Doing so, we find it hard for the algorithm to beat simpler ones based on recursive regressions and on the random walk in forecasting stock returns. We then argue that our results cast doubts on the plausibility of using learning classifier systems to represent agents process of expectations formation, an approach commonly found into the agent-based computational finance literature.
Keywords: genetic-based learning classifier systems, genetic algorithms, stock returns forecasting. JEL codes: D8, G1.
Galimberti, J.K. & Seabra, F. (2012) Conditioned export-led growth hypothesis: a panel threshold regressions approach, Revista de Economia, 38, p. 7-24.
Abstract: This paper deals with a reassessment of the export-led growth hypothesis on a panel threshold regressions context which allows testing for the existence of other variables conditioning the effects on the exports-growth nexus. The estimation covers a broad sample of 72 countries for the period 1974-2003. Overall, the empirical results give support to the export-led growth hypothesis, where the estimated thresholds indicate that growth was conditioned by countries initial levels of output and human capital. The effects of exports on growth, although exhibiting diminishing returns, were found to have great relevance in accelerating the process of income convergence across countries.
Keywords: export-led growth, panel threshold regressions, trade and growth. JEL codes: F43, O11, O40, O50.
Andrade, A.L.C. & Galimberti, J.K. (2011) Environmental Kuznets curve for CO2 emission: Analysis of a sample of OECD countries, Textos de Economia, 14, p. 82-102.
Abstract: This paper aims to verify the existence of a relationship between the stage of development in a sample of OECD countries and their respective levels of Carbon Dioxide (CO2) emissions in the period from 1971 to 2005. It tries to identify the stage in the development process which each country is within the framework of the environmental Kuznets curve. The results showed the phase in which each country is and allowed to classify them according to their environmental development. The evidence reveals that only 28% of the sample are in the downward phase, so it is possible to conclude that the effects of the process of technological development has not become widespread for most of the OECD economies.
Keywords: CO2 emissions, environmental Kuznets curve, economic growth.JEL codes: Q2, O40.
Suhadolnik, N. & Galimberti, J.K. & Da Silva, S. (2010) Robot traders can prevent extreme events in complex stock markets, Physica A: Statistical Mechanics and Its Applications, 389, p. 5182-5192.
Abstract: If stock markets are complex, monetary policy and even financial regulation may be useless to prevent bubbles and crashes. Here, we suggest the use of robot traders as an anti-bubble decoy. To make our case, we put forward a new stochastic cellular automata model that generates an emergent stock price dynamics as a result of the interaction between traders. After introducing socially integrated robot traders, the stock price dynamics can be controlled, so as to make the market more Gaussian.
Keywords: stock markets, robot traders, financial regulation, econophysics. JEL codes: G18, G01.
Galimberti, J.K. & Caldart, W.L. (2010) Exports and economic growth: analysis of Corede Serra's municipalities, Ensaios FEE, 31, p. 87-112.
Abstract: The paper aims to verify the existence of a relation between exports and economic growth in the cities belonging to the region of the Corede-Serra in the state of Rio Grande do Sul, Brazil, for the period of 1997 to 2004. It starts from the model formulated by Feder (1983), in which exports positively affect economic growth through two possible ways: positive externalities and resources reallocation to more productive activities. The results indicate that exports affect economic growth positively by resources reallocation from the non-export to the export sector, with productions factors being 62,65% more productive in the later. Therefore, it is deduced the importance of exports promotional policies, also in the municipal sphere, as a form to stimulate regional economic growth.
Keywords: economic growth, exports, regional economy.JEL codes: O41, F43, R11.
Abstract: This paper proposes a proxy-variable search procedure, based on a sensitivity analysis framework, aiming to provide a useful tool for the applied researcher whenever he faces measurement or proxy-variable uncertainties. Extending from the sensitivity analysis literature it proposes two main methodological innovations. The first relates to the usage of a proxies grouping process to obtain averaged coefficient estimators for theoretical explanatory variables that have more than one possible measure. The second is a proposal of using the actual empirical distribution of the available data to base the inference over the confidence probabilities in choosing each possible measure as proxy for a theoretical variable. This is done using the widely known bootstrapped residuals technique. Besides the methodological main focus, an empirical application is presented in the context of cross-country growth regressions. This empirical application provided favorable evidence to the neoclassical view about the specification of the human capital effect on growth. The results also emphasized how neglecting educational quality differentials might lead to wrong conclusions about the robustness of the relationship between human capital accumulation and economic growth.
Keywords: proxy-variable search, sensitivity analysis, estimation uncertainty. JEL codes: C1, O4.
Cheung, L., Galimberti, J. K. & Vermeulen, P. (2025). Evidence on the determinants and variation of idiosyncratic risk in housing markets.
Abstract: Using around one million repeat sales observations of single-family homes across New Zealand, over the period 1992 to 2021, we provide evidence that idiosyncratic risk in real house price appreciation varies considerably across houses. We find that idiosyncratic risk is time varying, depends negatively on the initial house price, varies strongly across locations and reduces significantly as the holding period of the house increases. Location is the most important of these factors. By buying an above the median house in a low-risk region, and holding on to the property for a longer period, households can significantly reduce idiosyncratic risk.
Keywords: idiosyncratic risk, house prices, housing markets . JEL codes: G1, R1.
Bajaro, D.F.E., Galimberti, J. K. & Qureshi, I. A. (2025). Monetary policy under fiscal stress: a forward-looking analysis of fiscal dominance. Draft available upon request.
Abstract: Subdued economic activity and low tax revenues, especially during crises, drive borrowing and increase public debt. Central banks may face pressure to deviate from policy targets during these periods to ease the debt burden. Under fiscal dominance, debt sustainability relies on low interest rates and high inflation rather than consolidation. This paper empirically tests the presence of fiscal dominance using forward-looking Taylor rules and data from 52 countries over three decades. The results detect fiscal dominance, with stronger effects in "de jure" inflation-targeting emerging economies with low central bank independence, especially those without debt rules and with high debt-to-GDP ratios. Emerging economies with high foreign currency-denominated debt are further affected by exchange rate debt valuation effects, and fiscal dominance leads their central banks to follow exchange rate stabilization policies. Since 2022--23, interest rate responses to fiscal imbalances have strengthened, posing challenges for future policy.
Keywords: fiscal dominance, monetary policy, Taylor rule, consensus forecasts. JEL codes: E31, E52, E58, E63.
Dalby, S., Gastwirth, J., Galimberti, J. K. & Wang, J. H. (2025). Gini variance estimation of grouped data with applications to night lights data. Draft available upon request.
Abstract: We propose a jackknife variance estimator for the Gini Index when observations stem from satellite images of lights on Earth at night and are estimated using a specific grouped data point estimator for the Gini Index. These estimators can be applied in any setting where the observations can be viewed as group sums of latent observations. We provide context for how our variance estimator fits into the wide literature on Gini Index variance estimation. This variance estimator is shown to be consistent for the asymptotic variance of the grouped data point estimator. We conduct a random effects modeling simulation to assess the robustness of both point and variance estimators to violations in underlying assumptions. We end with an application of this method to actual night lights data and summarize our findings in our conclusion.
Galimberti, J. K., Pichler, S. & Pleninger, R. (2024). Impact of natural disasters on light-based geospatial income inequality. Preliminary draft available upon request.
Abstract: We evaluate the impact of natural disasters on inequality, taking advantage of a newly constructed and here extended database of light-based geospatial income inequality (LGII) measures. Our new database provides Gini-coefficients for 233 countries and territories over the period from 1992 to 2019. The LGII is based on geo-located data of nighttime lights matched with gridded population counts, and is constructed to approximate income inequality from a benchmark measure that is not available for many countries affected by disasters. We estimate the impacts of disasters on inequality using panel regressions and a synthetic control group method (SCGM). The impact of natural disasters is hard to disentangle due to heterogeneity across disasters and the slowly moving nature of inequality. The panel regressions indicate that climatological disasters tend to decrease inequality, while meteorological disasters increase it. The SCGM analysis further illustrates the high variation of impacts across a selection of severe disasters. The magnitudes of the effects tend to decrease when looking at countries with multiple disasters, which suggests some degree of preparedness may condition the impacts of disasters on inequality.
Keywords: Inequality, disasters, nighttime lights, gridded population. JEL codes: D63, E01, O11, O57, Q54.
Chadwick, M., Cherry, R. & Galimberti, J. K. (2023). Non-response bias in household inflation expectations surveys.
Abstract: This paper uses micro-data from the Reserve Bank of New Zealand's Household Inflation Expectations survey to obtain a more accurate read of households' true inflation expectations by understanding how different demographic groups respond (or do not respond) to specific questions in the survey. Using a Heckman selection model, we assess whether there is item non-response bias in the survey by comparing the demographic characteristics of responders and non-responders. We quantify and demonstrate how to adjust for bias in aggregate (mean) measures of inflation expectations caused by item non-response. We show that there is a positive bias, and the aggregate inflation expectation series shifts down after the adjustment.
Applied Macroeconomics, AUT
Introduction to Econometrics, AUT
Coding for Economics Research, AUT
International Monetary Economics, ETH Zurich
Principles of Economics, ETH Zurich
Econometrics, University of Manchester