YouTube Video Review: Time Series Prediction by Siraj Ravel

One YouTuber whose content I really like is Siraj Ravel. Unfortunately he is not producing much content any longer and he sort of left YouTube under a cloud. I nevertheless like watching his old videos and maybe one day he will decide to go back to YouTube as a content creator.

One video that Siraj made, which I have found to be quite informative was on the subject of time series predictions:-

Eight different time series techniques that Siraj discussed in the video were:-

1. Naive approach
2. Simple average
3. Moving average
4. Weighted moving average
5. Simple exponential smoothing
6. Holt’s linear trend
7. Holt Winter’s method
8. Vector auto regression

Time series is a sequence of data points that measure the same thing over an ordered period of time. It is a series of numeric values, each with their own time stamp.

Time series datasets have three commonalities:-

1. New data is recorded as a new entry,
2. Data arrives in time order, and
3. Time is a primary axis.