I want to be able to indicate the effect of different factors on certain KPIs in a dashboard where the user has the option to select a factor (from a drop down linked to a variable) and set thresholds for classifying this as 'High', 'Medium' and 'Low', showing the average for weekly KPI value for High, Med and Low weeks.
The variables I have are:
$(vHigh) - above which classifies as a High week [linked to input box]
$(vLow) - below which classifies as a Low week [linked to input box]
(in between these two thresholds is a medium week)
$(vFactor) - selected from drop down
Data is at a weekly granularity. Something resembling:
LOAD * Inline [Week, KPI, Factor
1, 0.7, 17
2, 0.9, 18
3, 1.1, 14];
I figured I need to use a calculated dimension to classify each week as either high, medium or low based on the variables and then average the metric score for the week.
Can anyone help me with this?
For example, if I'm looking to determine the impact of time spent training on the speed at which a marathon runner covers 26miles each week.
The duration of training is the factor. I want to measure the average speed for the marathon across weeks when the level of training is above or below given thresholds.
The Thesholds are supposed to classify each week according to the factor.
I.e. Let's say I have five weeks where the time training is 11 hours, 9 hours, 6 hours and 2 hours and the High threshold is 7 with a Low threshold of 4
This means that the weeks 11 and 9 are High weeks, the 6 hour week is Medium and the 2 hour week is low. I then want to show the average KPI (Marathon time) for the high, medium and low weeks.