How to implement exponential smoothing for data without a trend/seasonality?
6 ビュー (過去 30 日間)
古いコメントを表示
Hi everyone,
My data is the form of AIS and I'm only focusing on Speed (SOG - Speed Over Ground) and Course (COG - Course Over Ground) parameters as a time series. I'm attaching a csv file that contains the time stamp (yyyymmddhhss) and SOG values for reference.
What I want to do is implement exponential smoothing to predict the next values. For the reference: the math behind this is as follows.
![](https://www.mathworks.com/matlabcentral/answers/uploaded_files/469887/image.png)
My question is:
Which exponential method will suit this the best? (I followed a bunch of articles and most of them say that simple exponential smoothing is ideal for data without trends/seasonality. But the results I got from that is not accurate and I would like to know if I can implement triple exponential method)
Is there a MATLAB method to implement this? (I found Holt Winters tool here, but it does not seem promising)
Also, is there a way to manually implement the above equations using MatLab? I'm not sure if I know how to carry out the first iteration :(
I'd appreciate any tip/recommendation. TIA!
0 件のコメント
回答 (0 件)
参考
カテゴリ
Help Center および File Exchange で Mathematics and Optimization についてさらに検索
Community Treasure Hunt
Find the treasures in MATLAB Central and discover how the community can help you!
Start Hunting!