You are developing a SQL Server Analysis Services (SSAS) cube for the accounts department. You create a measure group named Exchange Rate that consists of measures pertaining to currency exchange rate

Sum would add all exchange rate values across time periods together, producing a meaningless total rather than an average rate.

Count returns the number of records in the partition, not the average of the exchange rate values.

Min returns only the lowest exchange rate encountered, not the average.

Max returns only the highest exchange rate encountered, not the average.

DistinctCount counts unique values and does not compute an average.

None suppresses all aggregation, so the measure would not roll up to higher time granularities at all.

ByAccount uses the aggregation function defined on members of a linked account-type dimension, which is not applicable here without a matching account dimension configuration.

AverageOfChildren is a semi-additive aggregation function in SSAS that, when rolling up over time, returns the average of all child member values rather than their sum. For an exchange rate measure, summing rates across periods is semantically incorrect - an average rate is the meaningful value - so AverageOfChildren ensures that any rollup (e.g., from month to quarter) produces the correct average exchange rate over that interval.

FirstChild returns the value of the first child member in time - a snapshot behavior used for opening balances, not an average.

LastChild returns the value of the last child member in time - a snapshot behavior used for closing balances, not an average.

FirstNonEmpty returns the first child with a non-null value, which is a snapshot behavior not equivalent to averaging.