CA Performance Management - 3.7

PDF
20.2 3.7 3.6 3.5 3.2 3.1
English

Scorecard Projections

Scorecard projections use a customizable set of data points to predict future values for metrics. Projected values are calculated when the view is rendered, and are based on the historical time frame of the view. You can add the projected values to IM Custom View Group Scorecards.
capm370
HID_Scorecard_Projections
Scorecard projections use a customizable set of data points to predict future values for metrics. Projected values are calculated when the view is rendered, and are based on the historical time frame of the view. You can add the projected values to IM Custom View Group Scorecards.
Do not use scorecard projections for error metrics. Each error is a discrete event that is not affected by historical errors.
The scorecard view includes two methods to calculate projections:
  • Approximation
    The Approximation method uses the average from each time frame in the view to calculate the projection values.
    CA Performance Management
     calculates a least squares regression on the averages, then uses the line equation to project future values. This calculation method is faster than the Detailed Data method.
  • Detailed Data
    The Detailed Data method uses the polled data for the entire time frame of the view.
    CA Performance Management
     calculates a least squares regression for the entire set of data points. This calculation is more statistically accurate than the Approximation method, and provides extra columns in the view.
    Detailed data scorecard projections are supported only for gauge metrics (for example, Bits Out - Average Rate). Detailed data scorecard projections are
    not
     supported for counter metrics (for example, Bits Out - Total). Projection values are calculated on the As Polled (rate) data to ensure precision.
    These columns are hidden by default:
    • Slope
      Indicates the slope of the line equation.
    • Intercept
      Indicates the intercept of the line equation.
    • Degrees
      Degrees of freedom, which indicates the sample size.
    • Linear Fit
      Indicates the confidence level of the projected values as related to the sample data.
    • Days to Threshold
      Indicates the projected number of days before the specified critical threshold is reached.