Guillaume Chevillon
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Previous Organizations / Institutes
The organizations / institutes from the scholar's other articles in the past ten years
Research Subjects
Guillaume Chevillon has a total of 18 co-authors. Here are the top co-authors by number of shared publications:
Sophocles Mavroeidis
University of Oxford
3 shared publications- Alain Hecq
Department of Quantitative Economics Maastricht University The Netherlands
2 shared publications Christine Rifflart
2 shared publicationsAlain Hecq
1 shared publication- Anurag N. Banerjee
Durham University Business School Durham UK
1 shared publication Anurag Narayan Banerjee
1 shared publication
Latest Publications
- Open Access
- 28 July 2026
- Published by Elsevier in SSRN Electronic Journal
Testing Temperature Controllability: A Cointegration Approach to Climate Policy Evaluation
Controlling global temperatures requires suppressing an upward stochastic trend, a problem that falls naturally within the nonstationary control theory of Johansen and Juselius (2001, 2026). We exploit the Industrial Revolution as a natural experiment, constructing a dataset spanning AD 1000–2008 to estimate climate-economy relationships that are stable across this major regime shift. Our cointegrated VAR yields two equilibrium relations matching the structure of Stochastic DICE models: a damage equation linking output, consumption, and temperature (with estimated damage parameter ξ= 0.97, consistent with Nordhaus and Moffat, 2017), and a physical equilibrium between CO2 concentrations and temperature implying an equilibrium climate sensitivity in the range of Hassler et al. (2018). Parameter stability across the Industrial Revolution supports the invariance required for counterfactual policy analysis. Applying the Johansen-Juselius controllability condition, we reject non-controllability of temperature at conventional significance levels. Counterfactual simulations reveal sharply divergent costs: stabilizing temperatures at 1900 levels via output reduction (or degrowth) would have required foregoing 76% of 2008 GDP, whereas technology-based decarbonization (at an assumed zero cost) achieves the same target while increasing GDP (median gain of about 100%). This demonstrates that green investment can be self-financing and provides a toolbox for evaluating alternative climate policies.
- 28 July 2026
- Published by Elsevier in SSRN Electronic Journal
Open AccessTesting Temperature Controllability: A Cointegration Approach to Climate Policy Evaluation
Controlling global temperatures requires suppressing an upward stochastic trend, a problem that falls naturally within the nonstationary control theory of Johansen and Juselius (2001, 2026). We exploit the Industrial Revolution as a natural experiment, constructing a dataset spanning AD 1000–2008 to estimate climate-economy relationships that are stable across this major regime shift. Our cointegrated VAR yields two equilibrium relations matching the structure of Stochastic DICE models: a damage equation linking output, consumption, and temperature (with estimated damage parameter ξ= 0.97, consistent with Nordhaus and Moffat, 2017), and a physical equilibrium between CO2 concentrations and temperature implying an equilibrium climate sensitivity in the range of Hassler et al. (2018). Parameter stability across the Industrial Revolution supports the invariance required for counterfactual policy analysis. Applying the Johansen-Juselius controllability condition, we reject non-controllability of temperature at conventional significance levels. Counterfactual simulations reveal sharply divergent costs: stabilizing temperatures at 1900 levels via output reduction (or degrowth) would have required foregoing 76% of 2008 GDP, whereas technology-based decarbonization (at an assumed zero cost) achieves the same target while increasing GDP (median gain of about 100%). This demonstrates that green investment can be self-financing and provides a toolbox for evaluating alternative climate policies.
- Open Access
- 1 January 2024
- Published by Elsevier in SSRN Electronic Journal
The Queer Algorithm
This essay explores the intersection of queer theory and algorithmic design in the digital era. It emphasizes the potential of queering algorithms to render Artificial Intelligence more efficient at mapping more desirable possible futures. Drawing on multidisciplinary research, with an emphasis on social sciences and economics, I study to what extent the tools of Artificial Intelligence and data analysis currently incorporate in their design the diversity of cognitive and physical human experiences in space and time. I then argue that algorithmic practices rooted in deterministic, normative, or binary premises lack the robustness necessary for their joint purposes of modeling and predicting. I advocate for a systemic approach to reimagining and interacting with algorithms, highlighting the importance of maintaining long-term societal objectives, and of generating agency and autonomy in communities at the margin. This essay calls for a collaborative effort to redesign algorithms, making technology more reflective of and responsive to the complexities of human identity.
- 1 January 2024
- Published by Elsevier in SSRN Electronic Journal
Open AccessThe Queer Algorithm
This essay explores the intersection of queer theory and algorithmic design in the digital era. It emphasizes the potential of queering algorithms to render Artificial Intelligence more efficient at mapping more desirable possible futures. Drawing on multidisciplinary research, with an emphasis on social sciences and economics, I study to what extent the tools of Artificial Intelligence and data analysis currently incorporate in their design the diversity of cognitive and physical human experiences in space and time. I then argue that algorithmic practices rooted in deterministic, normative, or binary premises lack the robustness necessary for their joint purposes of modeling and predicting. I advocate for a systemic approach to reimagining and interacting with algorithms, highlighting the importance of maintaining long-term societal objectives, and of generating agency and autonomy in communities at the margin. This essay calls for a collaborative effort to redesign algorithms, making technology more reflective of and responsive to the complexities of human identity.
- 14 February 2020
- Published by Oxford University Press (OUP) in The Econometrics Journal
- Vol. 23 (2) , 297-315
Probabilistic forecasting of bubbles and flash crashes
We propose a near explosive random coefficient autoregressive model (NERC) to obtain predictive probabilities of the apparition and devolution of bubbles. The distribution of the autoregressive coefficient of this model is allowed to be centred at an O(T−α) distance of unity, with α ∈ (0, 1). When the expectation of the autoregressive coefficient lies on the explosive side of unity, the NERC helps to model the temporary explosiveness of time series and obtain related predictive probabilities. We study the asymptotic properties of the NERC and provide a procedure for inference on the parameters. In empirical illustrations, we estimate predictive probabilities of bubbles or flash crashes in financial asset prices.
- 14 February 2020
- Published by Oxford University Press (OUP) in The Econometrics Journal
- Vol. 23 (2) , 297-315
Probabilistic forecasting of bubbles and flash crashes
We propose a near explosive random coefficient autoregressive model (NERC) to obtain predictive probabilities of the apparition and devolution of bubbles. The distribution of the autoregressive coefficient of this model is allowed to be centred at an O(T−α) distance of unity, with α ∈ (0, 1). When the expectation of the autoregressive coefficient lies on the explosive side of unity, the NERC helps to model the temporary explosiveness of time series and obtain related predictive probabilities. We study the asymptotic properties of the NERC and provide a procedure for inference on the parameters. In empirical illustrations, we estimate predictive probabilities of bubbles or flash crashes in financial asset prices.
- Open Access
- 5 March 2019
- Published by Cambridge University Press (CUP) in Econometric Theory
- Vol. 36 (1) , 86-121
ROBUST INFERENCE IN STRUCTURAL VECTOR AUTOREGRESSIONS WITH LONG-RUN RESTRICTIONS
Long-run restrictions are a very popular method for identifying structural vector autoregressions, but they suffer from weak identification when the data is very persistent, i.e., when the highest autoregressive roots are near unity. Near unit roots introduce additional nuisance parameters and make standard weak-instrument-robust methods of inference inapplicable. We develop a method of inference that is robust to both weak identification and strong persistence. The method is based on a combination of the Anderson-Rubin test with instruments derived by filtering potentially nonstationary variables to make them near stationary using the IVX instrumentation method of Magdalinos and Phillips (2009). We apply our method to obtain robust confidence bands on impulse responses in two leading applications in the literature.
- 5 March 2019
- Published by Cambridge University Press (CUP) in Econometric Theory
- Vol. 36 (1) , 86-121
Open AccessROBUST INFERENCE IN STRUCTURAL VECTOR AUTOREGRESSIONS WITH LONG-RUN RESTRICTIONS
Long-run restrictions are a very popular method for identifying structural vector autoregressions, but they suffer from weak identification when the data is very persistent, i.e., when the highest autoregressive roots are near unity. Near unit roots introduce additional nuisance parameters and make standard weak-instrument-robust methods of inference inapplicable. We develop a method of inference that is robust to both weak identification and strong persistence. The method is based on a combination of the Anderson-Rubin test with instruments derived by filtering potentially nonstationary variables to make them near stationary using the IVX instrumentation method of Magdalinos and Phillips (2009). We apply our method to obtain robust confidence bands on impulse responses in two leading applications in the literature.
- 1 May 2018
- Published by Elsevier in Journal of Econometrics
- Vol. 204 (1) , 54-65
Generating univariate fractional integration within a large VAR(1)
- 1 May 2018
- Published by Elsevier in Journal of Econometrics
- Vol. 204 (1) , 54-65
Generating univariate fractional integration within a large VAR(1)
- 31 March 2018
- Published by Elsevier in Journal of Economic Dynamics and Control
- Vol. 90, 343-365
Perpetual learning and apparent long memory
- 31 March 2018
- Published by Elsevier in Journal of Economic Dynamics and Control
- Vol. 90, 343-365
Perpetual learning and apparent long memory
- Open Access
- 23 July 2017
- Published by Elsevier in SSRN Electronic Journal
Robustness of Multistep Forecasts and Predictive Regressions at Intermediate and Long Horizons
This paper studies the properties of multi-step projections, and forecasts that are obtained using either iterated or direct methods. The models considered are local asymptotic: they allow for a near unit root and a local to zero drift. We treat short, intermediate and long term forecasting by considering the horizon in relation to the observable sample size. We show the implication of our results for models of predictive regressions used in the financial literature. We show here that direct projection methods at intermediate and long horizons are robust to the potential misspecification of the serial correlation of the regression errors. We therefore recommend, for better global power in predictive regressions, a combination of test statistics with and without autocorrelation correction.
- 23 July 2017
- Published by Elsevier in SSRN Electronic Journal
Open AccessRobustness of Multistep Forecasts and Predictive Regressions at Intermediate and Long Horizons
This paper studies the properties of multi-step projections, and forecasts that are obtained using either iterated or direct methods. The models considered are local asymptotic: they allow for a near unit root and a local to zero drift. We treat short, intermediate and long term forecasting by considering the horizon in relation to the observable sample size. We show the implication of our results for models of predictive regressions used in the financial literature. We show here that direct projection methods at intermediate and long horizons are robust to the potential misspecification of the serial correlation of the regression errors. We therefore recommend, for better global power in predictive regressions, a combination of test statistics with and without autocorrelation correction.
- 1 May 2017
- Published by Elsevier in Journal of Econometrics
- Vol. 198 (1) , 1-9
Learning can generate long memory
- 1 May 2017
- Published by Elsevier in Journal of Econometrics
- Vol. 198 (1) , 1-9
Learning can generate long memory
- 1 January 2016
- Published by Elsevier in International Journal of Forecasting
- Vol. 32 (1) , 121-137
Multistep forecasting in the presence of location shifts
- 1 January 2016
- Published by Elsevier in International Journal of Forecasting
- Vol. 32 (1) , 121-137
Multistep forecasting in the presence of location shifts
- Open Access
- 7 May 2015
- Published by HAL
Long Memory Through Marginalization of Large Systems and Hidden Cross-Section Dependence
This paper shows that large dimensional vector autoregressive (VAR) models of finite order can generate long memory in the marginalized univariate series. We derive high-level assumptions under which the final equation representation of a VAR(1) leads to univariate fractional white noises and verify the validity of these assumptions for two specific models. We consider the implications of our findings for the variances of asset returns where the so-called golden-rule of realized variances states that they tend always to exhibit fractional integration of a degree close to 0:4.
- 7 May 2015
- Published by HAL
Open AccessLong Memory Through Marginalization of Large Systems and Hidden Cross-Section Dependence
This paper shows that large dimensional vector autoregressive (VAR) models of finite order can generate long memory in the marginalized univariate series. We derive high-level assumptions under which the final equation representation of a VAR(1) leads to univariate fractional white noises and verify the validity of these assumptions for two specific models. We consider the implications of our findings for the variances of asset returns where the so-called golden-rule of realized variances states that they tend always to exhibit fractional integration of a degree close to 0:4.