For local patterns that change quickly, these methods are preferred 1. data driven 2. model drive 3. trend and seasonality drive 4. all of the above

1. data driven

Explanation:

Data driven -

The pattern is localized, and changes quickly, So, the approach is analysed, on the given data in the real time.

Model driven

But, in case of model driven, the pattern of the data is not fixed.

In case of seasonality and trend driven,

in this method, there is every time a particular pattern in a fixed interval of time.

Hence, from the question, the most appropriate answer is data driven.