
218
Dran-View 6 User Guide
19.2. Weighted Statistics
The samples inside the current time limits are sorted in ascending sample
magnitude order along with their associated durations. A given sample is assumed
to be the best estimate of line conditions until a new sample comes along to replace
it. Therefore its duration (or weighting) is given as the time difference between the
sample and the next sample. Starting at the lowest sample magnitude the associated
sample durations are summed. When the duration sum equals or exceeds the
desired percentage of the total duration of all samples, the magnitude is extracted
and shown. These are the values used for the median and percentage value
estimates.
We can think of the result as:
For N% of the time the values are less than or equal to the N% value. For the rest of
the time it is greater.
The weighted standard deviation is carried out using the following formulas. D
denotes durations and X denotes sample magnitudes:
Duration of all samples = D
total
=
Where D
n
is equal to the time duration associated with each individual sample.
Weighted (by duration) Average of Samples =
weighted
=
Using the weighted average as our model the formula for the weighted standard
deviation is as follows:
Weighted (by duration) Standard Deviation (biased) =
Komentáře k této Příručce