univariate time series prediction with artificial neural network

1 次查看（过去 30 天）
Osman Yakubu
Osman Yakubu 2018-12-26
评论： Parvathy ravindranath 2020-12-25
I am new to MATLAB and time series and need help. I have a two column data of electricity consumption (Date, Consumption in kWd). I need a MATLAB code or procedure to enable me predict consumptions. I have 154 days of data and I want to prediction each consumption and plot it on a graph (actual, predicted) and calculate the root mean squared error. Thanks.
  3 个评论
显示 1更早的评论
Osman Yakubu
Osman Yakubu 2018-12-27
编辑：Osman Yakubu 2018-12-28
Thanks for your reply, it was really helpful. How do I plot the predicted and actual values? I also need the RMSE.
Please help.
Thanks
Kevin Chng
Kevin Chng 2019-1-4
编辑：madhan ravi 2019-1-4
Sorry for my late reply,
(Actual - Predicted) % Errors
(Actual - Predicted).^2 % Squared Error
mean((Actual - Predicted).^2) % Mean Squared Error
RMSE = sqrt(mean((Actual - Predicted).^2)); % Root Mean Squared Error

请先登录，再进行评论。

请先登录，再回答此问题。

采纳的回答

Kevin Chng
Kevin Chng 2019-1-4
refer to the link : https://www.mathworks.com/help/deeplearning/ref/narnet.html. Replace the dataset with your dataset.
For Calculating RMSE,
RMSE = sqrt(mean((Actual - Predicted).^2));

更多回答（0 个）

类别

在 Help Center 和 File Exchange 中查找有关 Sequence and Numeric Feature Data Workflows 的更多信息

标签

Community Treasure Hunt

Find the treasures in MATLAB Central and discover how the community can help you!

Start Hunting!

Translated by