White noise is a special type of time series where data doesn’t follow any pattern. It is a sequence of random data where every value has a time series associated with it.
Since no pattern is followed, it is difficult to predict white noise.
For data to be white noise, it should follow the below conditions:
- Constant Mean = 0
- Constant Standard Deviation
- No Autocorrelation
Autocorrelation is the measure of how correlated a series is with the past version of itself. No autocorrelation means there is no clear correlation between the present and past values of a series.