| Name: | Description: | Size: | Format: | |
|---|---|---|---|---|
| 1009.41 KB | Adobe PDF |
Authors
Advisor(s)
Abstract(s)
Neste relatório apresentam-se resultados de um estudo estatístico que procura contribuir
para um melhor entendimento da problemática inerente à liberalização do setor
elétrico em Portugal e dos desafios que esta liberalização, existente desde meados de
2007, trás aos seus intervenientes.
Iniciam-se os trabalhos com um estudo que pretende avaliar a existência de relação
entre o Preço de Mercado da eletricidade e um conjunto de variáveis potencialmente
explicativas/condicionantes do Preço de Mercado. Neste estudo consideram-se duas
abordagens. A primeira usa a função de correlação cruzada para avaliar a existência de
relação do tipo linear entre pares de variáveis. A segunda considera o teste causalidade
de Granger na avaliação de uma relação de causa e efeito entre esses pares.
Este estudo avaliou a relação entre o Preço de Mercado da eletricidade e 19 variáveis
ditas condicionantes distribuídas por três categorias distintas (consumo e produção de
eletricidade; indicadores climáticos; e energias primárias). O intervalo de tempo em
estudo cinge-se ao biénio 2012-2103. Durante este período avaliam-se as relações entre
as variáveis em diversos sub-períodos de tempo em ciclos de consumo representativos
do consumo em baixa (fim de semana) e de consumo mais elevado (fora de vazio)
com os valores observados de cada uma das variáveis tratados com uma base horária
e diária (média).
Os resultados obtidos mostram a existência relação linear entre algumas das variáveis
em estudo e o preço da eletricidade em regime de mercado liberalizado, mas raramente
é possível identificar precedência temporal entre as variáveis.
Considerando os resultados da análise de correlação e causalidade, apresenta-se ainda
um modelo de previsão do Preço de Mercado para o curto e médio prazo em horas de
período fora de vazio.
Within this report, results of a statistical study are presented, which seek to contribute to a better understanding of the inherent problems existing since mid-2007 regarding the liberalization of the electrical sector in Portugal and the challenges that this liberalization has brought its stakeholders. To begin the report, a study was performed with the aim of evaluating the existence of a relationship between the Price of Electricity and a set of potentially explanatory variables/constraints of the electricity price. In this study two approaches were considered. The first uses the cross-correlation method to evaluate the existence of the linear type relationship between variable pairs. The second considers the Granger causality test in the assessment of a cause-effect relationship between these pairs. This study evaluated the relationship between the Market Price of electricity and 19 variables stated constraints distributed into three distinct categories (electricity consumption and production, climate indicators, and primary energy). The time interval under study was confined to the biennium 2012-2103. During this period, the relationship between variables were evaluated in different sub-periods of time in cycles representative of low consumer consumption (weekend) and higher consumption (peak time) with each of the variables values observed treated on an hourly and daily basis (the average). The results show the existence of a linear relationship between some of the variables under study and the price of electricity under a liberalized market, but it is rarely possible to identify temporal precedence between variables. Considering the results of the correlation and causality analysis, a forecasting model for the Market Price is presented for the short medium term for hour of peak time period.
Within this report, results of a statistical study are presented, which seek to contribute to a better understanding of the inherent problems existing since mid-2007 regarding the liberalization of the electrical sector in Portugal and the challenges that this liberalization has brought its stakeholders. To begin the report, a study was performed with the aim of evaluating the existence of a relationship between the Price of Electricity and a set of potentially explanatory variables/constraints of the electricity price. In this study two approaches were considered. The first uses the cross-correlation method to evaluate the existence of the linear type relationship between variable pairs. The second considers the Granger causality test in the assessment of a cause-effect relationship between these pairs. This study evaluated the relationship between the Market Price of electricity and 19 variables stated constraints distributed into three distinct categories (electricity consumption and production, climate indicators, and primary energy). The time interval under study was confined to the biennium 2012-2103. During this period, the relationship between variables were evaluated in different sub-periods of time in cycles representative of low consumer consumption (weekend) and higher consumption (peak time) with each of the variables values observed treated on an hourly and daily basis (the average). The results show the existence of a linear relationship between some of the variables under study and the price of electricity under a liberalized market, but it is rarely possible to identify temporal precedence between variables. Considering the results of the correlation and causality analysis, a forecasting model for the Market Price is presented for the short medium term for hour of peak time period.
Description
Keywords
Correlação Causalidade MIBEL Preço da electricidade Correlation Causality Electricity Price
Citation
Publisher
Instituto Politécnico do Porto. Instituto Superior de Engenharia do Porto