Forecasting and risk analysis for Vietnam industrial full market price (FMP) / spot price

Loading...
Thumbnail Image

URL

Journal Title

Journal ISSN

Volume Title

School of Business | Master's thesis

Authors

Department

Major/Subject

Mcode

Language

en

Pages

62

Series

Abstract

As Vietnam transitions toward a competitive wholesale electricity market under the Net Zero 2050 directive, the introduction of Decree 57/2025/ND-CP and the Direct Power Purchase Agreement (DPPA) has altered the risk landscape for energy participants. This thesis investigates the stochastic modeling and risk quantification of the Vietnam electricity spot price (Full Market Price - FMP) using a newly released five-year dataset (2020–2024) from the National Power System and Market Operation Company (NSMO). The research evaluates three distinct time-series forecasting frameworks: Seasonal Autoregressive Integrated Moving Average (SARIMA), Extreme Gradient Boosting (XGBoost), and Long Short-Term Memory (LSTM) networks. The results suggest that univariate models, trained on historical prices, consistently outperformed multivariate models incorporating exogenous weather and fuel variables. These findings provide corporate offtakers and generators with a validated quantitative tool for managing basis risk and structuring Contracts for Difference (CfD) within Vietnam’s evolving energy market.

Description

Supervisor

Leppänen, Ilkka

Thesis advisor

Ho, Tung
Stuart, Marc

Other note

Citation