Shijie Deng is an associate professor in ISyE. He received a B.S. in applied mathematics from Peking University in P.R. China, an M.S. in mathematics from the University of Minnesota, and a Ph.D. in industrial engineering and operations research from the University of California.
Dr. Deng's research interests include financial asset pricing and real options valuation, financial engineering applications in energy markets, electricity transmission pricing and auction design, risk management and contract theory in supply chains, stochastic modeling and simulation. He has consulted with several private and public companies on issues of pricing, risk management, and asset valuation in the deregulated electricity industry.
Published in:
Publication year: 1999
The restructuring of electric power industry has become a global trend. As reforms to the electricity supply industry spread rapidly across countries and states, many political and economical issues arise as a result of people debating over which approach to adopt in restructuring the vertically integrated electricity industry. This dissertation addresses issues of transmission pricing, electricity spot price modeling, as well as risk management and asset valuation in a competitive electricity industry.
A major concern in the restructuring of the electricity industries is the design of a transmission pricing scheme that will ensure open-access to the transmission networks. I propose a priority-pricing scheme for zonal access to the electric power grid that is uniform across all buses in each zone. The Independent System Operator (ISO) charges bulk power traders a per unit ex ante transmission access fee based on the expected option value of the generated power with respect to the random zonal spot prices. The zonal access fee depends on the injection zone and a self-selected strike price determining the scheduling priority of the transaction. Inter zonal transactions are charged (or cred-ited) with an additional ex post congestion fee that equals the zonal spot price di erence. The unit access fee entitles a bulk power trader to either physical injection of one unit of energy or a compensation payment that equals to the difference between the realized zonal spot price and the selected strike price. The ISO manages congestion so as to minimize net compensation payments and thus, curtailment probabilities corresponding to a particular strike price may vary by bus. I calculate the rational expectation equilibria for several numerical examples and demonstrate that the eciency losses of the proposed second best scheme relative to the effcient dispatch solutions are modest.
The rest of the dissertation deals with the issues of modeling electricity spot prices, pricing electricity financial instruments and the corresponding risk management applications. The aforementioned global trend of restructuring opens the electricity wholesale markets worldwide. Modeling the spot prices of electricity is important for the market participants who need to understand the risk factors in pricing electricity financial instruments such as electricity forwards, options and cross-commodity derivatives. It is also essential for the analysis of nancial risk management, asset valuation, and project financing.
In the setting of di usion processes with multiple types of jumps, I examine three mean-reversion models for modeling the electricity spot prices. I impose some structure on the coeffcients of the di usion processes, which allows me to easily compute the prices of contingent claims (or, financial instruments) on electricity by Fourier methods. The jump-diffusion models which I propose have the features of mean-reversion, stochastic volatility, and regime switching. I derive the pricing formulas for various electricity derivatives and examine how the prices vary with di erent modeling assumptions.
I demonstrate a couple of risk management applications of the electricity financial instruments. I also construct a real options approach to value electric power generation and transmission assets both with and without accounting for the operating characteristics of the assets such as start-up cost, ramp-up time and variable heat rate. The implications of the mean-reversion jump-di usion models on financial risk management and real asset valuation in competitive electricity markets are illustrated. With a discrete trinomial lattice modeling the underlying commodity prices, I estimate the effects of operational characteristics on the asset valuation by means of numerical examples that incorporate these aspects using stochastic dynamic programming.
The restructuring of electric power industry has become a global trend. As reforms to the electricity supply industry spread rapidly across countries and states, many political and economical issues arise as a result of people debating over which approach to adopt in restructuring the vertically integrated electricity industry. This dissertation addresses issues of transmission pricing, electricity spot price modeling, as well as risk management and asset valuation in a competitive electricity industry.
A major concern in the restructuring of the electricity industries is the design of a transmission pricing scheme that will ensure open-access to the transmission networks. I propose a priority-pricing scheme for zonal access to the electric power grid that is uniform across all buses in each zone. The Independent System Operator (ISO) charges bulk power traders a per unit ex ante transmission access fee based on the expected option value of the generated power with respect to the random zonal spot prices. The zonal access fee depends on the injection zone and a self-selected strike price determining the scheduling priority of the transaction. Inter zonal transactions are charged (or cred-ited) with an additional ex post congestion fee that equals the zonal spot price di erence. The unit access fee entitles a bulk power trader to either physical injection of one unit of energy or a compensation payment that equals to the difference between the realized zonal spot price and the selected strike price. The ISO manages congestion so as to minimize net compensation payments and thus, curtailment probabilities corresponding to a particular strike price may vary by bus. I calculate the rational expectation equilibria for several numerical examples and demonstrate that the eciency losses of the proposed second best scheme relative to the effcient dispatch solutions are modest.
The rest of the dissertation deals with the issues of modeling electricity spot prices, pricing electricity financial instruments and the corresponding risk management applications. The aforementioned global trend of restructuring opens the electricity wholesale markets worldwide. Modeling the spot prices of electricity is important for the market participants who need to understand the risk factors in pricing electricity financial instruments such as electricity forwards, options and cross-commodity derivatives. It is also essential for the analysis of nancial risk management, asset valuation, and project financing.
In the setting of di usion processes with multiple types of jumps, I examine three mean-reversion models for modeling the electricity spot prices. I impose some structure on the coeffcients of the di usion processes, which allows me to easily compute the prices of contingent claims (or, financial instruments) on electricity by Fourier methods. The jump-diffusion models which I propose have the features of mean-reversion, stochastic volatility, and regime switching. I derive the pricing formulas for various electricity derivatives and examine how the prices vary with di erent modeling assumptions.
I demonstrate a couple of risk management applications of the electricity financial instruments. I also construct a real options approach to value electric power generation and transmission assets both with and without accounting for the operating characteristics of the assets such as start-up cost, ramp-up time and variable heat rate. The implications of the mean-reversion jump-di usion models on financial risk management and real asset valuation in competitive electricity markets are illustrated. With a discrete trinomial lattice modeling the underlying commodity prices, I estimate the effects of operational characteristics on the asset valuation by means of numerical examples that incorporate these aspects using stochastic dynamic programming.