Mitsuru Igami
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
Mitsuru Igami has a total of 1 co-authors. Here are the top co-authors by number of shared publications:
Takuo Sugaya
Stanford Graduate School of Business
1 shared publication
Latest Publications
- 18 August 2021
- Published by Oxford University Press (OUP) in Review of Economic Studies
- Vol. 89 (3) , 1460-1494
Measuring the Incentive to Collude: The Vitamin Cartels, 1990–99
Do mergers help or hinder collusion? This article studies the stability of the vitamin cartels in the 1990s and presents a repeated-games approach to quantify “coordinated effects” of a merger. We use data and direct evidence from American courts and European agencies to show the collusive incentive of the short-lived vitamin C cartel was likely to be negative when it actually collapsed in 1995, whereas the incentives of the long-lived cartels (vitamins A and E, and beta carotene) were unambiguously positive until the prosecution in 1999. Simulations suggest some mergers could have prolonged the vitamin C cartel, but others could have further destabilized it, because both the direction and magnitude of coordinated effects depend not only on the number of firms but also on their cost asymmetry.
- 18 August 2021
- Published by Oxford University Press (OUP) in Review of Economic Studies
- Vol. 89 (3) , 1460-1494
Measuring the Incentive to Collude: The Vitamin Cartels, 1990–99
Do mergers help or hinder collusion? This article studies the stability of the vitamin cartels in the 1990s and presents a repeated-games approach to quantify “coordinated effects” of a merger. We use data and direct evidence from American courts and European agencies to show the collusive incentive of the short-lived vitamin C cartel was likely to be negative when it actually collapsed in 1995, whereas the incentives of the long-lived cartels (vitamins A and E, and beta carotene) were unambiguously positive until the prosecution in 1999. Simulations suggest some mergers could have prolonged the vitamin C cartel, but others could have further destabilized it, because both the direction and magnitude of coordinated effects depend not only on the number of firms but also on their cost asymmetry.
- Open Access
Artificial Intelligence as Structural Estimation: Economic Interpretations of Deep Blue, Bonanza, and AlphaGo
Artificial intelligence (AI) has achieved superhuman performance in a growing number of tasks, but understanding and explaining AI remain challenging. This paper clarifies the connections between machine-learning algorithms to develop AIs and the econometrics of dynamic structural models through the case studies of three famous game AIs. Chess-playing Deep Blue is a calibrated value function, whereas shogi-playing Bonanza is an estimated value function via Rust's (1987) nested fixed-point method. AlphaGo's "supervised-learning policy network" is a deep neural network implementation of Hotz and Miller's (1993) conditional choice probability estimation; its "reinforcement-learning value network" is equivalent to Hotz, Miller, Sanders, and Smith's (1994) conditional choice simulation method. Relaxing these AIs' implicit econometric assumptions would improve their structural interpretability.
Open AccessArtificial Intelligence as Structural Estimation: Economic Interpretations of Deep Blue, Bonanza, and AlphaGo
Artificial intelligence (AI) has achieved superhuman performance in a growing number of tasks, but understanding and explaining AI remain challenging. This paper clarifies the connections between machine-learning algorithms to develop AIs and the econometrics of dynamic structural models through the case studies of three famous game AIs. Chess-playing Deep Blue is a calibrated value function, whereas shogi-playing Bonanza is an estimated value function via Rust's (1987) nested fixed-point method. AlphaGo's "supervised-learning policy network" is a deep neural network implementation of Hotz and Miller's (1993) conditional choice probability estimation; its "reinforcement-learning value network" is equivalent to Hotz, Miller, Sanders, and Smith's (1994) conditional choice simulation method. Relaxing these AIs' implicit econometric assumptions would improve their structural interpretability.
- Open Access
- 1 January 2011
- Published by Springer Nature in Review of Industrial Organization
- Vol. 38 (1) , 1-21
Does Big Drive Out Small?
- 1 January 2011
- Published by Springer Nature in Review of Industrial Organization
- Vol. 38 (1) , 1-21
Open AccessDoes Big Drive Out Small?