Research On Long-term And Dynamic Relationship Between Factors Of Power Grid Investment Base On Time Series Analysis

Nowadays state grid company is called to be on-mission for safe and reliable supply of electricity. Electricity supply affects not only national economy but also political security and social stability. State grid companies have invested heavily in grid construction and upgrading in recent years. But it is facing problems about financial pressure and lower return on assets since the generation price keeps rising and sales price is controlled by the government. This paper analyzes factors of power grid investment from two perspectives (macro-economic and electricity markets) to help the company timely grasp the rule of power gird investment, which would be of great theoretical and practical meanings to the scientific continuing development of the company.Power grid investment involves many complex factors. The paper describes factor from macroeconomic perspective and electricity markets perspective such as GDP, electricity comsumption and price, and so on. Then base on the statistical data of US from year1997to2006, the paper studies the elastic coefficient between investment and its factors. And on this basis it builds the co-integration model that aims at the long-run equilibrium relationship among power grid investment, economy and electricity price and examines the model. In the mean time, the paper also applies error correction theory to set up short-term fluctuations adjusting modle. The results of numeric examples show that co-integration model can analyze proper adjustment range of investment on the basis of long-term equilibrium relationship. The error correction model reveals the influence of power grid investment to short-term fluctuations and the regulation of non-equilibrium processes to balance process.Power grid investment path is the running performance of business cycle in electricity market. This paper empirically analyzes the long-term trends and cyclical trends of power grid investment and its factors on the basis of Hodrick-Prescott theory. The results reflect that power grid investment, electricity consumption and GDP have the same primary period under the same standards, and the major period is supposed to be nine to eleven years based on "Peak-Peak" method and five percentage point of volatility skew. After that the paper emploies the recently developed vector autoregression to examine the dynamic relationship between power grid investment and it factors. Impulse response function shows the response characteristics of power grid investment to different factors of short-term disturbances. Variance decomposition reveals the contribution that structural impact of different variables to the power grid fluctuations.