Lag selection in unit root tests: a literature review
The econometric analysis may be one of the most common ways to model and forecast different time series problems, such as the electricity demand. In this type of analysis, the presence of unit root may lead to unreliable forecasts. Hence, the correct identification of the presence of unit root on th...
| Autores: | , , , , |
|---|---|
| Tipo de documento: | artigo |
| Estado: | Versão publicada |
| Data de publicação: | 2022 |
| País: | Brasil |
| Recursos: | Centro Universitário de Belo Horizonte (UNIBH) |
| Repositório: | Revista e-xacta |
| Idioma: | português |
| OAI Identifier: | oai:ojs.periodicos.uninove.br:article/22061 |
| Acesso em linha: | https://periodicos.uninove.br/exacta/article/view/22061 |
| Access Level: | Acceso aberto |
| Palavra-chave: | unit root lag selection electricity demand stationarity raiz unitária seleção de defasagens demanda por eletricidade estacionariedade |
| Resumo: | The econometric analysis may be one of the most common ways to model and forecast different time series problems, such as the electricity demand. In this type of analysis, the presence of unit root may lead to unreliable forecasts. Hence, the correct identification of the presence of unit root on the series to be modeled is essential. In order to perform this task, unit root tests, such as ADF, can be applied. One of key steps in this test procedure is to properly select the number of lags to be used. In this paper, we present in both quantitative and qualitative ways, that research on this subject is in expansion and there is no consensus on how to select the number of lags to be applied in the test. Therefore, it is evident that this is a subject in which further research is needed. |
|---|