Basic Measures

Unknown macro: {scrollbar}

The property pane for a basic measure includes the following tabs:

Description
General
Aggregation
Scope

The Description tab

Property	Description
Name	Enter the name of the measure level.
Description	Enter a description of the level.
Message when measure is out of scope	The message specified here is shown in the Browser and Portal to inform users that they are trying to query the measure in a scope where it is not defined.
Not visible for users	When this option is selected, the measure will not be available in the Browser. This option is valuable in early stages of the project when tests need to be conducted, allowing developers to add measures and hide them so that users can't see them. Datamarts need to be published in the server for this feature to work. Datamarts opened locally will display all the measures regardless of this setting.

The General tab

Property	Description
Value	Choose the sources field containing the value for each record's measure.
Store Value As	For every measure you can select between Double (64 bits) or Integer (32 bits) if the measure only stores integer numbers.
Scale Factor	This is an additional optimization. If you have floating point numbers, you can select to store them as Integers by using a Scale Factor. For example: the number 1234,567890... can be stored as an Integer as 1234567 with a Scale Factor of 3. O3 knows that this integer number is actually 1234,567 because the Scale Factor is 3 and manage it in this way. The range of numbers you can manage with an Integer depends on the scale factor you use.
Measure Format	Opens the measure Format dialog. See Formatting Numbers

The Aggregation tab

Property	Description
For Aggregation Use	Indicates the operation used to aggregate values for higher levels in the dimension. See Aggregation Modes
Special Date Aggregation	Enables a special aggregation mode based on Dates. See Aggregation Modes
For Weighted Average Use	Define which field is used for the weighted average. The dropdown list includes only fields defined as type integer.
Default Value	Default value for the measure when no data is specified. Options are NaN or zero.

The Scope tab

A measure's range permits to specify the dimensions and levels for which the measure can be queried (i.e. it is defined) and the entry level or base starting from this data.
If we take for example, a forecast measure in a cube with Date (Year, Quarter, Month), Product (Family, Line, Product), Clients (Client type, Client) and Salespersons (Salesperson) dimensions, the measure is not usually available in all the dimensions or in all the levels.

Generally there isn't a divided forecast on each product's level (the measure will not be defined on this level), but it will only be available on a family o maybe line level. The forecast divided in salespersons probably won't be available, so the measure will not be completely defined for this dimension.

Finally, if we analyze the date dimension, unlike other measure (where the data comes with lower levels of detail and then are aggregated to higher levels), the forecast can have Year of Quarter as entry level.

The definition of a measure's range is performed by specifying the behavior for each one of the cube's dimensions in the table of ranges. It's important to consider that the definition of the measure's entry levels has to be consistent with the data sources' definitions.
There are 3 main types of behavior in each dimension:

Standard
Undefined
Manual

Standard

This is the behavior more commonly used within the cubes. The entry level of the data matches the lowest level (or highest detail) of the hierarchy and all the levels above this one are calculated according to the aggregation level defined for the measure. In this case the measure is defined for all the dimension levels.

Undefined

This behavior permits to declare that the measure is not defined for this dimension.

Manual

In this case the behavior is declared level by level by the user.

The user will specify which is the entry level for the dimension, which levels will be obtained through aggregation from the one immediately below and in which levels the measure will not be defined.

Ther first line corresponds to the root level. This level is very important since it provides a way to indicate that the elements of the fisrt level may not aggregate (totals may not make sense there)

For instance, suppose there is a dimension called "Currency" that represents the currency used to buy or sell products. This dimension would have just one level with the different currencies that the company uses.

Also suppose that there is a measure called "Amount" that represents the amount of each transaction. Each amount should be associated o a particular currency.

Clearly, any analysis that does not filter a specific currency would be incorrect since it would add amounts represented in different currencies.

To avoid this situation, it is possible to indicate that the "Amount" measure can not be aggregated (summed in this case) for the root of the "Currency" dimension. This way O3 will not show any result and notify the user if no currency is selected.

Possible options for the scopes are:

	Input	Values in this level are taken directly from the data sources. Underneath this level there should only be undefined or constant levels.
	Aggregation	Values in this level are calculated using the aggregation model defined for the measure.
	Undefined	The measure is undefined for this level
	Constant	Values in this level are all the same and equal to their parent in the hierarchy.

Note
There should be only one Input level for each dimension.
Aggregation levels should always be on top of input levels or aggregation levels.

When a measure is marked as undefined for a certain level in a dimension, it means that it makes no sense to aggregate the measure for that level.

This restriction also applies to predefined calculated elements such as Sum and Totals.

An example of this would be a dimension with only one level that for a certain measure has its root marked as undefined.

When analyzing the cube, the measure could be located in the X axis and the dimension in the Y axis. A row added to the table as the Sum of the elements would make no sense.

When the cube is published in the O3 Server this situation is detected and the measure will be shown as "- -".

Unknown macro: {scrollbar}