Econometric modeler. Forecasting a SARIMA model
13 visualizaciones (últimos 30 días)
Mostrar comentarios más antiguos
Maria Perez-Tabernero
el 23 de Abr. de 2020
Editada: Maria Perez-Tabernero
el 23 de Abr. de 2020
Hello!
I have data from the first 3 weeks of March regarding electricity prices (for every hour during the first 21 days of the month, there is a specific price). I would like to model the time series as a SARIMA model with the Econometric Modeler App. As price changes depending on the day of the week and the hour of the day, I would like to include that. I have a table with two columns: date for every price (HH-dd-MM-yy) and the correspondent price.
Until now, I have:
- Modelled SARIMA_Value with the Econometric Modeler App. I think it automatically fits the model according to the given parameters p, d, q, P, D, Q. I have included m=3 because there are 3 weeks in my data. However, when importing the data and plotting it, the graph shows the prices but not the dates, the x axis is "Index", but it should be "Date".

2. Used estimate(SARIMA_Value, values_from_first_3_weeks)
I have the following questions:
- How can I include the dates into the Econometric App?
- Am I estimating the model properly?
- How can I forecast the prices for the fourth week?
Thank you!
0 comentarios
Respuestas (0)
Ver también
Categorías
Más información sobre Conditional Mean Models en Help Center y File Exchange.
Community Treasure Hunt
Find the treasures in MATLAB Central and discover how the community can help you!
Start Hunting!