Open Loop forecasting vs closed loop forecasting

Hi, I am currently using LSTM for time series forecasting. I am actually confused between Open Loop forecasting vs closed loop forecasting. Can someone explain it more thoroughly?

Honestly not sure which forecasting to use for my scenario. Just want to understand the difference and when i should be using each type.

Answer 1

[Zuber]

Hi Justin,

As per my understanding of the question, you want to know the difference between open loop and closed loop forecasting. Based on MathWorks documentation regarding the same:

• Open loop forecasting is used when you have the actual data available. For example, suppose you have true values of a time series data for time steps ‘1’ to ‘t-1’ and you want to predict the value at instant ‘t+1’. In such a scenario, you should record the true value at time step ‘t’ and use it as an input to make prediction for time step ‘t+1’.
• Closed loop forecasting predicts subsequent time steps in a sequence by using the previous predictions as input. In this case, the model does not require the true values to make the prediction. For example, suppose you want to predict the value for time steps ‘t’ through ‘t+k’ of the sequence using data collected in time steps ‘1’ through ‘t−1’ only. To make predictions for time step i, use the predicted value for time step i−1 as input. The closed loop forecasting is used when one wants to forecast multiple subsequent time steps or when you do not have true values to provide to the network before making the next prediction.

For more information on this topic, please refer to the following example that shows how to forecast time series data using a long short-term memory (LSTM) network: https://www.mathworks.com/help/deeplearning/ug/time-series-forecasting-using-deep-learning.html

I hope this answers your query.