Quantcast
Channel: SQL Server Analysis Services forum
Viewing all articles
Browse latest Browse all 14337

Simple problem on the prediction precision.

$
0
0

Simple problem on the prediction precision.

For all articles, I would like to calculate the maximum value of prognosis and outcome.
For all articles, I would like to calculate the absolute difference between the forecast and actual.

This shall always be done for the selected dimension, for example, customer.

But if, for example, the dimension group of customers used to be every single article summarized first and then the absolute difference / maximum calculation.

Is this possible to solve as a calculated member?

Customer1

Forecast

Actual

Max

Abs Delta

Precision

Article1

90

50

90

40

55,56%

Article2

50

100

100

50

50,00%

Sum

140

150

190

90

52,63%

Customer2

Forecast

Actual

Max

Abs Delta

Precision

Article1

50

90

90

40

55,56%

Article2

50

100

100

50

50,00%

Sum

100

190

190

90

52,63%

Customer1 +Customer2

Forecast

Actuall

Max

Abs Delta

Precision

Article1

140

140

140

0

100,00%

Article2

100

200

200

100

50,00%

Sum

240

340

340

100

70,59%

Customer1 +Customer2

Forecast

Actual

Max

Abs Delta

Precision

Article1

=B13+B6

=C13+C6

=MAX(B20:C20)

=ABS(B20-C20)

=1-E20/D20

Article2

=B14+B7

=C14+C7

=MAX(B21:C21)

=ABS(B21-C21)

=1-E21/D21

Sum

=SUMMA(B20:B21)

=SUMMA(C20:C21)

=SUMMA(D20:D21)

=SUMMA(E20:E21)

=1-E22/D22


Viewing all articles
Browse latest Browse all 14337

Trending Articles



<script src="https://jsc.adskeeper.com/r/s/rssing.com.1596347.js" async> </script>