Hello, I was started work with HTM Studio two days ago and I have a question.
Can you explain me what is the difference between aggregation methods average and sum?
Why sum method find less anomalies than average method?
And when what methods are better to use?
Maybe this questions are very easy, but I don’t have any programmer skills and I try to understand it
Hi @Alisa, let’s say you have three datapoints: [6, 7, 9], and we want to aggregate those three down to a single point.
Using the Average (‘mean’) setting, our new aggregated datapoint would be: . The values in our newly aggregated data will closely match those of the original data, even though we will be missing values.
Using the Sum setting, the new datapoint will be: . This new value is far from any of our original values, but the overall shape of the new data may be helpful.
As usual, “it depends on your data and what you want to do with it”.