Can Neural Network Toolbox's time series model use data that are observed with different time-lags?

1 次查看（过去 30 天）
SANG GYU LEE
SANG GYU LEE 2017-3-11
回答： Greg Heath 2017-3-12
Hi, I am thinking about using the Neural Network Toolbox for fitting a time-series neural network model as written the link below.
Neural Network Time-Series Prediction and Modeling https://www.mathworks.com/help/nnet/gs/neural-network-time-series-prediction-and-modeling.html
Regarding this, I have two questions. Thank you in advance for your time.
1. However, my data shows sometimes 2 months lags and 3 month lags (or sometimes 1 month lag). Can I use this raw data to fit a model? or Do I need an additional approach? What could it be?
2. My input variables might not be independent. Would that bring a problem? Is this model smart enough to consider all the dependencies by itself?

请先登录，再回答此问题。

回答（1 个）

Greg Heath
Greg Heath 2017-3-12
1. There is no problem if you can interpolate the series at equidistant points.
2. Correlated inputs are not a problem unless you want to rank their their importance.
Hope this helps.
Thank you for formally accepting my answer
Greg

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

产品

Community Treasure Hunt

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

Start Hunting!

Translated by