The system identification toolbox comes with the command pem which can construct a state spcae model through "prediction error minimization". sounds a lot like your problem, and a similar example is given in the help.
Time Series Forecasting
1 次查看(过去 30 天)
显示 更早的评论
There is a data from 4 to 40. ie Inflow Jan 1975- 32 mcm Feb 1975- 24 mcm .... Dec 1995 - 8 mcm Now i want it to train it and forecast for 1995-1997.it should be 3*5*1 ( input, hidden, output). Inputs should be Inflow(n-1), Inflow(n-2), Inflow(n-3).I have tried all the threads . A code will be very much helpful. Thanks
0 个评论
回答(1 个)
另请参阅
类别
在 Help Center 和 File Exchange 中查找有关 Data Preparation Basics 的更多信息
Community Treasure Hunt
Find the treasures in MATLAB Central and discover how the community can help you!
Start Hunting!