Time Series Revisited
I have been asked a couple of times for a script that would reproduce the results in the time series forecasting series. I finally managed to do it. In the process I found out that a couple of the queries needed to be tuned:
- In the airline example described in Part 2, the normalization shift and scale parameters were computed using the whole data. A better methodology would be to use only the training data for computing these parameters. This alternative normalization scheme also impacts the results for the airline example described in Part 3 as the latter builds on the transformations described on Part 2
- The results for the electric load example in Part 3 did not reflect the forecasting query. They reflected an earlier version of the query. The correct results are a little better
I have also made some edits to the Part 2 and Part 3 posts to reflect the points described above.
Labels: Time Series
I really like your time series article. It would be better more that some one should take the lead in providing good examples on forecasting and data mining, specially taking real time problems.
Posted by Anonymous | 7/19/2007 01:53:00 AM
Thanks Adil. I tried to provide some examples on real data in the third post in the series (link). The second example, specially, comes from a competition. If you have any interesting examples you would like to share I would happy to write a post on it.
Posted by Marcos | 11/15/2007 03:09:00 PM