I remove the top percentile of ppq when drawing graphs, because outliers which are often errors (e.g. quantity/items confusion) make reading the chart harder.
However, quite often this means we miss interesting data.
Do some surveys of which data gets clipped and reconsider.
I remove the top percentile of ppq when drawing graphs, because outliers which are often errors (e.g. quantity/items confusion) make reading the chart harder.
However, quite often this means we miss interesting data.
Do some surveys of which data gets clipped and reconsider.