关键词: 光伏发电, 条件异方差, 动态规划法, 最优决策

Abstract: The key problem for photovoltaic investment decisions is whether there is a reliable method that can be applied to assess the optimal critical value and the maximum opportunity value. This study choses 836 day closing prices (from January 4, 2012 to June 24, 2015) of a stock, which has a “twinborn” relationship with the value of photovoltaic investment program, to establish a volatility forecasting model. On this basis, this paper revises the dynamic programming method of the investment decision-making. Then, the optimal critical value, the maximum opportunity value and the trend of these two values under the different volatility are given. This research shows that: as a result of the conditional heteroscedasticity of the “twinborn” stock price, the optimal critical value and the biggest opportunity value have different sensitivity for different volatility—When the volatility increases, these two values increase, but the increasing extent are different; When the volatility reaches a certain value, the increasing extent of the two values are obviously improved. Therefore, the analysis of the volatility helps to improve the quality of decision-making, and reduce the loss of the enterprises.

Key words: photovoltaic power, conditional heteroscedasticity, dynamic programming method, optimal decision