Katsunobu Sasanuma
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
Katsunobu Sasanuma has a total of 12 co-authors. Here are the top co-authors by number of shared publications:
- Alan Scheller-Wolf
Tepper School of Business, Carnegie Mellon University, Pittsburgh, Pennsylvania 15213
5 shared publications - Christine Pitocco
College of Business, Stony Brook University, Stony Brook, New York
2 shared publications - Mohammad Delasay
College of Business, Stony Brook University, Stony Brook, New York 11794
2 shared publications - Thomas R. Sexton
Stony Brook University
2 shared publications Akira Hibiki
1 shared publicationAlan Scheller-Wolf
1 shared publication
Latest Publications
- Open Access
- 20 October 2026
- Published by Elsevier in SSRN Electronic Journal
Algorithmic Inference and Insider Trading Enforcement: Hidden Markov Models, Market Surveillance, and the Gap Between Detection and Proof
Hidden Markov Models ("HMMs") and other probabilistic methods can support statistical inferences about otherwise unobservable states from trading behavior. Such models are therefore attractive tools for insider-trading surveillance, where regulators can observe trades but cannot directly observe what a trader knew at the time of trading. Yet a trading based inference consistent with an "informed state" is not itself legal proof of insider trading. This Article examines how HMM-based inference relates to the substantive requirements of U.S. insider-trading law. It argues that the legal significance of a trading-based inference depends on the substantive standard against which the inference is evaluated. Under a use standard, an HMM inference remains relatively remote from liability because the model does not establish that inside information caused or influenced the trade. Under Rule 10b5-1's awareness standard, however, the inference is conceptually closer to the relevant legal inquiry when other evidence connects the trading pattern to the trader's informational condition at the time of trading. Even then, the two are not equivalent: HMM-based inference by itself does not establish awareness of particular material nonpublic information, nor does it establish other elements such as materiality, nonpublic status, and duty. This Article therefore proposes a human-in-the-loop surveillance framework combining HMM inference with conditional-entropy analysis. It treats the framework’s outputs as probabilistic investigative signals that can help regulators identify suspicious trading, prioritize investigations, and determine what additional evidence is necessary to establish liability. A brief comparison with Japanese insider-trading law further shows that although HMM technology itself is jurisdiction-neutral, its legal significance is not. The usefulness of the same HMM inference depends on whether a jurisdiction requires actual use or awareness of material nonpublic information and what exceptions or defenses accompany that standard.
- 20 October 2026
- Published by Elsevier in SSRN Electronic Journal
Open AccessAlgorithmic Inference and Insider Trading Enforcement: Hidden Markov Models, Market Surveillance, and the Gap Between Detection and Proof
Hidden Markov Models ("HMMs") and other probabilistic methods can support statistical inferences about otherwise unobservable states from trading behavior. Such models are therefore attractive tools for insider-trading surveillance, where regulators can observe trades but cannot directly observe what a trader knew at the time of trading. Yet a trading based inference consistent with an "informed state" is not itself legal proof of insider trading. This Article examines how HMM-based inference relates to the substantive requirements of U.S. insider-trading law. It argues that the legal significance of a trading-based inference depends on the substantive standard against which the inference is evaluated. Under a use standard, an HMM inference remains relatively remote from liability because the model does not establish that inside information caused or influenced the trade. Under Rule 10b5-1's awareness standard, however, the inference is conceptually closer to the relevant legal inquiry when other evidence connects the trading pattern to the trader's informational condition at the time of trading. Even then, the two are not equivalent: HMM-based inference by itself does not establish awareness of particular material nonpublic information, nor does it establish other elements such as materiality, nonpublic status, and duty. This Article therefore proposes a human-in-the-loop surveillance framework combining HMM inference with conditional-entropy analysis. It treats the framework’s outputs as probabilistic investigative signals that can help regulators identify suspicious trading, prioritize investigations, and determine what additional evidence is necessary to establish liability. A brief comparison with Japanese insider-trading law further shows that although HMM technology itself is jurisdiction-neutral, its legal significance is not. The usefulness of the same HMM inference depends on whether a jurisdiction requires actual use or awareness of material nonpublic information and what exceptions or defenses accompany that standard.
- Open Access
- 17 April 2026
- Published by Elsevier in SSRN Electronic Journal
Mitigating the Optimizer’s Curse in Forecast-Driven Battery Arbitrage via Conservative Guard Bands
When point price forecasts are fed directly into a deterministic battery scheduler, forecast noise creates spurious arbitrage opportunities and the optimizer tends to select cycles whose ex ante value is overstated. This paper proposes a lightweight robustification method that constructs conservative buy- and sell-side price inputs from rolling empirical quantiles of forecast errors (guard bands) and embeds them in a standard day-ahead dynamic-programming scheduler without scenario generation. A stylized threshold model shows that the optimal conservative buffer increases with forecast noise and with the effective profitability threshold, implying greater value of skepticism when round-trip efficiency is low. The method is evaluated using Japan Electric Power Exchange (JEPX) spot prices (November 2023 to October 2025) for four areas in Japan, two forecast models, and four efficiency settings. Conservative guard bands improve realized performance mainly by suppressing bad trades. Moderate conservativeness raises mean profit and risk-adjusted return in low-efficiency settings, while its value weakens as efficiency rises. Even at high efficiency, however, very mild guard bands can sometimes improve risk-adjusted performance by pruning marginal noise-driven cycles. The analysis also reveals strong regional heterogeneity, with PV-rich Kyushu exhibiting the highest value capture. Overall, the value of conservativeness depends on battery efficiency, forecast quality, and market structure.
- 17 April 2026
- Published by Elsevier in SSRN Electronic Journal
Open AccessMitigating the Optimizer’s Curse in Forecast-Driven Battery Arbitrage via Conservative Guard Bands
When point price forecasts are fed directly into a deterministic battery scheduler, forecast noise creates spurious arbitrage opportunities and the optimizer tends to select cycles whose ex ante value is overstated. This paper proposes a lightweight robustification method that constructs conservative buy- and sell-side price inputs from rolling empirical quantiles of forecast errors (guard bands) and embeds them in a standard day-ahead dynamic-programming scheduler without scenario generation. A stylized threshold model shows that the optimal conservative buffer increases with forecast noise and with the effective profitability threshold, implying greater value of skepticism when round-trip efficiency is low. The method is evaluated using Japan Electric Power Exchange (JEPX) spot prices (November 2023 to October 2025) for four areas in Japan, two forecast models, and four efficiency settings. Conservative guard bands improve realized performance mainly by suppressing bad trades. Moderate conservativeness raises mean profit and risk-adjusted return in low-efficiency settings, while its value weakens as efficiency rises. Even at high efficiency, however, very mild guard bands can sometimes improve risk-adjusted performance by pruning marginal noise-driven cycles. The analysis also reveals strong regional heterogeneity, with PV-rich Kyushu exhibiting the highest value capture. Overall, the value of conservativeness depends on battery efficiency, forecast quality, and market structure.
- Open Access
- 1 January 2024
- Published by IGI Global Scientific Publishing in International Journal of Technology and Human Interaction
- Vol. 20 (1) , 1-23
Evaluating the Effectiveness of Recommendation Engines on Customer Experience Across Product Categories
Artificial intelligence (AI)-powered tools such as recommendation engines are widely used in online marketing and e-commerce; however, online retailers often deploy these tools without understanding which human factors play a role in which products and at which stage of the customer journey. Understanding the interaction between AI-powered tools and humans can help practitioners create more effective online marketing platforms and improve human interaction with e-commerce tools. This paper examines customers' reliance on recommendation engines when purchasing fashion goods, electronics, and media content such as video and music. This paper also discusses the potential for improvement in recommendation engines in online marketing and e-commerce.
- 1 January 2024
- Published by IGI Global Scientific Publishing in International Journal of Technology and Human Interaction
- Vol. 20 (1) , 1-23
Open AccessEvaluating the Effectiveness of Recommendation Engines on Customer Experience Across Product Categories
Artificial intelligence (AI)-powered tools such as recommendation engines are widely used in online marketing and e-commerce; however, online retailers often deploy these tools without understanding which human factors play a role in which products and at which stage of the customer journey. Understanding the interaction between AI-powered tools and humans can help practitioners create more effective online marketing platforms and improve human interaction with e-commerce tools. This paper examines customers' reliance on recommendation engines when purchasing fashion goods, electronics, and media content such as video and music. This paper also discusses the potential for improvement in recommendation engines in online marketing and e-commerce.
- Open Access
- 25 March 2022
- Published by Elsevier in Operations Research Perspectives
- Vol. 9, 100230
A marginal analysis framework to incorporate the externality effect of ordering perishables
- 25 March 2022
- Published by Elsevier in Operations Research Perspectives
- Vol. 9, 100230
Open AccessA marginal analysis framework to incorporate the externality effect of ordering perishables
- Open Access
- 1 March 2022
- Published by Elsevier in Results in Control and Optimization
- Vol. 6
Controlling arrival and service rates to reduce sensitivity of queueing systems with customer abandonment
- 1 March 2022
- Published by Elsevier in Results in Control and Optimization
- Vol. 6
Open AccessControlling arrival and service rates to reduce sensitivity of queueing systems with customer abandonment
- Open Access
- 14 January 2022
- Published by Elsevier in Operations Research Perspectives
- Vol. 9
An opaque selling scheme to reduce shortage and wastage in perishable inventory systems
- 14 January 2022
- Published by Elsevier in Operations Research Perspectives
- Vol. 9
Open AccessAn opaque selling scheme to reduce shortage and wastage in perishable inventory systems
- Open Access
- 8 November 2021
- Published by MDPI AG in Smart Cities
- Vol. 4 (4) , 1391-1402
Evaluating the Dynamic Impact of Theater Performances and Sports Events on Parking Demand in Downtown Pittsburgh
The number of drivers using parking facilities (parking demand) in downtown Pittsburgh is highly variable throughout business operating hours, which makes an efficient operation of parking facilities challenging and results in congestion around the facilities. In this study, we applied an event-based ordinary least squares (OLS) regression model to the parking data set provided from one of the parking facilities, the Theater Square Garage in downtown Pittsburgh. We demonstrated that our model achieved a high R-squared value during time periods when parking demand is highly variable. Using the model, we revealed the dynamic (time-dependent) impact of theater performances and sports events on parking demand. This dynamic information can help facility managers appropriately adjust their operating settings (e.g., the number of staff and fee structure) during surge or vacant time periods accordingly. This model is applicable to various businesses in downtown areas that have increased customer flow from theater performances and sports events, not only parking garages.
- 8 November 2021
- Published by MDPI AG in Smart Cities
- Vol. 4 (4) , 1391-1402
Open AccessEvaluating the Dynamic Impact of Theater Performances and Sports Events on Parking Demand in Downtown Pittsburgh
The number of drivers using parking facilities (parking demand) in downtown Pittsburgh is highly variable throughout business operating hours, which makes an efficient operation of parking facilities challenging and results in congestion around the facilities. In this study, we applied an event-based ordinary least squares (OLS) regression model to the parking data set provided from one of the parking facilities, the Theater Square Garage in downtown Pittsburgh. We demonstrated that our model achieved a high R-squared value during time periods when parking demand is highly variable. Using the model, we revealed the dynamic (time-dependent) impact of theater performances and sports events on parking demand. This dynamic information can help facility managers appropriately adjust their operating settings (e.g., the number of staff and fee structure) during surge or vacant time periods accordingly. This model is applicable to various businesses in downtown areas that have increased customer flow from theater performances and sports events, not only parking garages.
- Open Access
- 7 September 2021
- Published by MDPI AG in Mathematics
- Vol. 9 (18) , 2187
Asymptotic Analysis for Systems with Deferred Abandonment
This short paper concerns the analysis of the M/M/k queueing system with customer abandonment. In this system, service managers provide a finite buffer space, which is a waiting area that prevents customers from abandoning the system. Abandonment of the system can occur from reneging (exiting from the queue while waiting), and/or balking (leaving the system without waiting). We derive an analytical expression to represent the impact of the buffer space capacity on the delay probability and the abandonment probability for a system with deferred abandonment. The result indicates the provision of the buffer space in a large system could only increase the delay probability while the abandonment probability remains unchanged. Despite the benevolent intentions of service managers, providing a buffer space may exacerbate the performance of larger systems.
- 7 September 2021
- Published by MDPI AG in Mathematics
- Vol. 9 (18) , 2187
Open AccessAsymptotic Analysis for Systems with Deferred Abandonment
This short paper concerns the analysis of the M/M/k queueing system with customer abandonment. In this system, service managers provide a finite buffer space, which is a waiting area that prevents customers from abandoning the system. Abandonment of the system can occur from reneging (exiting from the queue while waiting), and/or balking (leaving the system without waiting). We derive an analytical expression to represent the impact of the buffer space capacity on the delay probability and the abandonment probability for a system with deferred abandonment. The result indicates the provision of the buffer space in a large system could only increase the delay probability while the abandonment probability remains unchanged. Despite the benevolent intentions of service managers, providing a buffer space may exacerbate the performance of larger systems.
- Open Access
A Marginal Analysis Framework to Incorporate the Externality Effect of Ordering Perishables
Finding the optimal policy for multi-period perishable inventory systems requires solving computationally-expensive stochastic dynamic programs (DP). To avoid the difficulty of solving DP models, we propose a framework that uses an externality term to capture the long-term impact of ordering decisions on the average cost over an infinite horizon. By approximating the externality term, we yield a tractable approximate optimality condition, which is solved through standard marginal analysis. The resulted policy is near-optimal in long-run average cost and ordering decisions.
Open AccessA Marginal Analysis Framework to Incorporate the Externality Effect of Ordering Perishables
Finding the optimal policy for multi-period perishable inventory systems requires solving computationally-expensive stochastic dynamic programs (DP). To avoid the difficulty of solving DP models, we propose a framework that uses an externality term to capture the long-term impact of ordering decisions on the average cost over an infinite horizon. By approximating the externality term, we yield a tractable approximate optimality condition, which is solved through standard marginal analysis. The resulted policy is near-optimal in long-run average cost and ordering decisions.
- Open Access
Markov Chain Decomposition Based On Total Expectation Theorem
A divide-and-conquer approach to analyzing Markov chains (MCs) is not utilized as widely as it could be, despite its potential benefits. One primary reason for this is the fact that most MC decomposition approaches involve a complex and inflexible methodology: decomposed subchains must be disjoint, transition rates of these decomposed subchains must be altered in a way tailored to the particular MC model, and the procedure to aggregate suchains needs to incorporate a nonlinear normalization constraint, complicating the analytical expression of performance measures. In contrast, we propose a versatile yet simple decomposition method for continuous time MCs based on the total expectation theorem. Leveraging the properties of this theorem, our method has great flexibility in the choice of subchains, and the procedure to obtain expected values of interest is simply a linear summation of subchains' properties, which is not affected by the normalization constraint. We prove that to maintain the correct distribution of decomposed subchains one may use our novel termination scheme, a modification of transition rates, that ensures partial flow conservation at boundary states. This termination scheme is applicable to MCs with any structure, since the scheme depends only on the boundary-state distribution, not on the structure of the MCs. To demonstrate the generality and capability of our method, we analytically solve various models, such as a congestion-based staffing queue and a Markov-modulated Mt/Mt/1 queue. As not all systems admit an analytical solution, we complement this analysis with numerical studies of MCs with various sizes using the algorithm based on our method.
Open AccessMarkov Chain Decomposition Based On Total Expectation Theorem
A divide-and-conquer approach to analyzing Markov chains (MCs) is not utilized as widely as it could be, despite its potential benefits. One primary reason for this is the fact that most MC decomposition approaches involve a complex and inflexible methodology: decomposed subchains must be disjoint, transition rates of these decomposed subchains must be altered in a way tailored to the particular MC model, and the procedure to aggregate suchains needs to incorporate a nonlinear normalization constraint, complicating the analytical expression of performance measures. In contrast, we propose a versatile yet simple decomposition method for continuous time MCs based on the total expectation theorem. Leveraging the properties of this theorem, our method has great flexibility in the choice of subchains, and the procedure to obtain expected values of interest is simply a linear summation of subchains' properties, which is not affected by the normalization constraint. We prove that to maintain the correct distribution of decomposed subchains one may use our novel termination scheme, a modification of transition rates, that ensures partial flow conservation at boundary states. This termination scheme is applicable to MCs with any structure, since the scheme depends only on the boundary-state distribution, not on the structure of the MCs. To demonstrate the generality and capability of our method, we analytically solve various models, such as a congestion-based staffing queue and a Markov-modulated Mt/Mt/1 queue. As not all systems admit an analytical solution, we complement this analysis with numerical studies of MCs with various sizes using the algorithm based on our method.