asked 1 day ago
Hello,
We are looking into adding an aggregation type which would calculate a weighted average based on another column's weight values, instead of calculating a straight average as is currently supported by Flexmonster. I am attaching a screenshot of a table which shows what I'm referring to. i.e. A would be a sum of A1 and A2 weighted by their corresponding values in the Weights column, divided by the total weight.
So the calculation for A would be (1000 * 0.3 + 2000 * 0.6)/0.9 = 1,666.67.
Is there a way of doing that currently? If so, could you give us some pointers on how to achieve it? If not, could Flexmonster implement it? Thank you.
⋅ 17 hours ago
Hello, Will!
Thank you for writing to us.
Such a formula can be created with Flexmonster's calculated values - a feature for creating custom formulas based on existing measures. Assuming that your dataset contains individual fields with values, weights, and string hierarchies for "A" and "B", you can create a formula as is shown in this example: https://jsfiddle.net/flexmonster/0tw5f3xh/
Here is an explanation of this example:
"value_x_weight" - values multiplied by weights. For correct output, this formula should multiply values individually, which can be flagged with individual: true. Also, since this is a part of the final formula, so it should be hidden with the active: false parameter."weighted_avg" - the formula used for final visualization, which divides the "value_x_weight" on the sum of weights.We are looking forward to hearing your feedback.
Best Regards,
Maksym