I'm working on time series data and forecasting using weighted moving average. I have done the brute force analysis of the data and found a solution and also developed a technique to find the optimal window size. However, my problem is what are the current benchmark techniques for deciding the window size in time series analysis using weighted moving average
Eiman Kanjo
I personally try to avoid fixed size window..instead I use an adaptive window which changes with data. However, deciding on the size of window should be experimental..try different sizes and then choose the one which gives you higher accuracy.
0 votes
0 thanks
Nirav Bhatt
Window size is also depends on arrival pattern of input data and processing semantic used.
0 votes
0 thanks
- Badges
- Science topic
- Similar topics
- Books
More Ankita Sinha's questions See All
Similar questions and discussions