A Discoverer calculation examples

About ranking

Ranking functions compute the ranked list position of an item with respect to other items in an ordered list.

Examples:

• "Example: Assign ranks to sales figures"

• "Example: Assign ranks to sales figures within region"

• "Example: Show the top three selling cities per region"

• "Example: Show the top three and bottom three selling cities per region"

Hint: You can also use analytic function templates to quickly and easily create calculations based on ranking functions (for more information, see "How to create a new calculation using an analytic function template".

Example: Assign ranks to sales figures

This example calculates the ranked list position of a set of sales figures.

Figure A-4 Worksheet containing the Rank Sales calculation

Description of "Figure A-4 Worksheet containing the Rank Sales calculation"

The worksheet shows the ranked list position of sales figures for cities in the year 2000.

Notes

• By default, ranked results data is sorted in ascending order (ASC), nulls first (NULLS FIRST). The additional DESC parameter sorts the results in descending order, which ranks the highest value with the Rank 1.

Example: Assign ranks to sales figures within region

This example calculates the ranked list position of a set of sales figures within each region.

Figure A-5 Worksheet containing the Rank sales within Region calculation

Description of "Figure A-5 Worksheet containing the Rank sales within Region calculation"

The worksheet shows the ranked list position of sales figures for cities grouped by region within year.

Notes

• By default, ranked results data is sorted in ascending order (ASC), nulls first (NULLS FIRST). The additional DESC parameter sorts the results in descending order, which ranks the highest value with the Rank 1.

Example: Show the top three selling cities per region

This example calculates the ranked list position of a set of sales figures and displays the top three selling cities.

Figure A-6 Worksheet containing the Rank Top calculation used in a condition

Description of "Figure A-6 Worksheet containing the Rank Top calculation used in a condition"

The worksheet shows a ranked list of the top three highest sales figures for each region within year.

Notes

• By default, ranked results data is sorted in ascending order (ASC), nulls first (NULLS FIRST). The additional DESC parameter sorts the results in descending order, which ranks the highest value with the Rank 1.

• Hint: To quickly filter the list to the first, second, or third ranked cities, pivot the Rank Top item to the page axis.

Example: Show the top three and bottom three selling cities per region

This example calculates the ranked list position of a set of sales figures and displays the top three and bottom three performing cities per region.

Figure A-7 Worksheet containing the Rank Top calculation used in a condition

Description of "Figure A-7 Worksheet containing the Rank Top calculation used in a condition"

The worksheet shows a ranked list of the three highest and three lowest sales figures for each region within year.

Notes

• This analysis involves three steps:

  1. Assign ranks to Cities on Sales SUM in descending order, as Rank Top.

  2. Assign ranks to Cities on Sales SUM in ascending order, as Rank Bottom.

  3. Displaying only Rank Top, filter the data using a Condition to return only the top three and bottom three ranked Cities.

• In the example, in the 'Central' Region, the top three cities are ranked 1, 2, and 3; the bottom three cities are ranked 5, 6, and 7. In the 'East' Region, the top three cities are ranked 1, 2, and 3; the bottom three cities are ranked 6, 7, and 8, and so on.

About banding

Banding is a type of ranking that divides a list of values in a partition into a specified number of groups called bands (also known as buckets) and assigns each value to a band.

Examples:

• "Example: Banding by value (1)"

• "Example: Banding by value (2)"

• "Example: Banding by rank"

Hint: You can also use analytic function templates to quickly and easily create calculations based on banding functions (for more information, see "How to create a new calculation using an analytic function template".

Two common types of banding are:

• Banding by value - this divides values into groups according to their value (also known as equi-width bands). This type of analysis is also known as frequency distribution.

  Here, the function typically takes the largest value minus the lowest value, and divides the result by the number of bands required. This value defines the range of each Band.

  Values are then assigned to bands according to which range they fall into. Therefore, the number of values in each Band might differ. For example, if we have 100 values and divide them into four equi-width bands, each band might contain different numbers of values.

Figure A-8 Banding By value

Description of "Figure A-8 Banding By value"

Use the GREATEST function or the CASE function to produce equi-width bands based on value.

Note: If that database version that you are using does not support IF or CASE statements, you can use DECODE instead. Alternatively, you can create a PL/SQL function in the database.

Hint: You can also use the WIDTH_BUCKET function to produce equi-width bands (see "Example: Producing equi-width bands using WIDTH_BUCKET").

• Banding by rank - this divides values into groups according to their rank (also known as equi-height bands). This type of analysis is also known as percentile analysis (e.g. 4 bands give quartiles).

  Here, the function divides the number of values in the partition by the number of bands, which gives the number of values in each band. An equal number of values are then placed in each band. For example, if we have 100 values and divide them into four equi-height bands, each band contains 25 values.

Figure A-9 Banding By Rank

Description of "Figure A-9 Banding By Rank"

Use the NTILE function to produce equi-height bands based on rank.

Example: Banding by value (1)

This example divides sales figures into bands according to their value (e.g. for frequency analysis). For more information, see "Example: Banding by value (2)".

Figure A-10 Worksheet containing the Sales Bands calculation

Description of "Figure A-10 Worksheet containing the Sales Bands calculation"

The worksheet shows equi-width bands for sales figures for cities in the central region within year.

Notes

• Using the Central Region and Year 2000 as an example, this function takes the largest value (45,758) minus the smallest value (7,749) and divides it by four ((45,758-7,749)/4), giving four equal Bands of 9,502.25. This gives four bands with the following ranges:

  • Band 1 - 36,255.75 to 45,758

  • Band 2 - 26,753.5 to 36,255.75

  • Band 3 - 17,251.25 to 26,753.5

  • Band 4 - 7,749 to 17,251.25

• Each value is placed in one of the four Bands depending on which range the Sales SUM value falls into.

• The FLOOR function returns the largest integer equal to or less than n. For example, in Dallas, the expression FLOOR(Sales SUM-Min Sales for Region) returns the smallest integer value from 7,749 minus 7,749, which returns 0. When used in conjunction with the GREATEST function (see calculation above), the expression GREATEST(1,4-FLOOR((Sales SUM-Min Sales for Region) returns the largest value from either 1 or, 4 minus the smallest integer value from 7,749 minus 7,749 (4 minus 0 equals 4). In other words, the expression returns the value 4.

• You can also use the WIDTH_BUCKET function to produce equi-width bands (see "Example: Producing equi-width bands using WIDTH_BUCKET").

Example: Banding by value (2)

This example divides sales figures into bands according to their value. The example creates the same results as the example in "Example: Banding by value (1)", except that it uses a CASE statement rather than the GREATEST function (e.g. for frequency analysis).

Figure A-11 Worksheet containing the Sales Bands 2 calculation

Description of "Figure A-11 Worksheet containing the Sales Bands 2 calculation"

The worksheet shows equi-width bands for sales figures for cities in the central region within year.

Notes

• This function uses a series of IF statements in the form of a CASE function to assign sales figures into bands (see Band ranges below).

• You can also use the WIDTH_BUCKET function to produce equi-width bands (see "Example: Producing equi-width bands using WIDTH_BUCKET").

Example: Banding by rank

This example divides sales figures into two bands according to their rank (e.g. for percentile analysis).

Figure A-12 Worksheet containing the Sales Bands 3 calculation

Description of "Figure A-12 Worksheet containing the Sales Bands 3 calculation"

The worksheet shows equi-height bands for sales figures for cities in the central region within year.

Notes

• Using the Central Region and Year 2000 as an example, this function takes the number of values (which is six) and divides it by two, giving three values per Band. It then takes the list of values ordered by Sales SUM and places values one, two, and three in band 1, values four, five, and six in band 2.

About windowing

Windowing functions are used to compute aggregates using values from other rows. For example, cumulative, moving, and centered aggregates.

Note: For more information about using row-based and time-based intervals, see "Examples of using row-based and time-based intervals".

Examples:

• "Example: Calculate a three month moving sales average"

• "Example: Show the cumulative values of sales"

• "Example: Compare sales figures across time using windowing"

Hint: You can also use analytic function templates to quickly and easily create calculations based on windowing functions (for more information, see "How to create a new calculation using an analytic function template".

Two common types of windowing are:

• Windowing with time-based intervals - here, a time-based interval is based on a value relative to an existing value. For example, three months preceding a date value.

  Note: Time-based intervals are also known as logical offsets.

  Time-based intervals are useful when data has missing rows. For example, you might not have seven rows for sales figures for seven days in a week, but you still want to find a weekly average.

  For example, if we have a list of monthly sales figures, a logical window might compute a moving average of the previous three months (inclusive of the current month). When calculating the average, the calculation assumes a NULL value for months missing from the list. In the example below, the three-month moving average for November assumes NULL values for the missing months September and October.

Figure A-13 Using a time-based interval

Description of "Figure A-13 Using a time-based interval"

• Windowing with row-based intervals - here, a row-based interval is based on a value that is a specified number of rows from an existing value. For example, three rows from the current item.

  Note: Row-based intervals are also known as physical offsets.

  Row-based intervals are useful when data does not have missing rows. For example, if you know that you have seven rows for sales figures for seven days in a week, you can calculate a weekly average by starting the interval six rows before the current row (inclusive).

  For example, if we have a list of monthly sales figures, a physical window might compute a moving average of the previous three rows. When calculating the average, the calculation ignores months missing from the list. In the example below, the three-month moving average for November uses June, July, and November in the calculation.

Figure A-14 Using a row-based interval

Description of "Figure A-14 Using a row-based interval"

Note: For more information about using row-based and time-based intervals, see "Examples of using row-based and time-based intervals".

Example: Calculate a three month moving sales average

This example uses a time-based interval to calculate a moving three month sales average.

Note: Moving averages are also known as rolling averages.

Figure A-15 Worksheet containing the Moving Average calculation

Description of "Figure A-15 Worksheet containing the Moving Average calculation"

The worksheet shows a moving three month average for sales figures for months in the year 2000.

Notes

• Note that you define the RANGE INTERVAL as '2', not '3', even though you want a three month window. This is because the window expression implicitly includes the current row. Therefore, in this example, the INTERVAL '2' plus the current row gives a total of three months (2 + current row = 3).

Example: Show the cumulative values of sales

This example uses a row-based interval to calculate the cumulative value of sales.

Figure A-16 Worksheet containing the Cumulative Total calculation

Description of "Figure A-16 Worksheet containing the Cumulative Total calculation"

The worksheet shows a cumulative total for sales figures for cities in the central region.

Example: Compare sales figures across time using windowing

This example uses a time-based interval to calculate sales figures for previous years. This enables you to compare sales figures over different years, or compare previous years' sales figures with other values, such as spending in previous years.

Figure A-17 Worksheet containing the Sales Last Year calculation

Description of "Figure A-17 Worksheet containing the Sales Last Year calculation"

For each row, the worksheet shows the sales total for the previous year.

Notes

• In the example above, the Sales Last Year value for 1998 is NULL because the database does not contain information for 1997.

• You can also use LAG/LEAD functions to compare values across time (see "LAG/LEAD function examples".

About reporting functions

Reporting functions are used to compute aggregates.

Examples:

• "Example: Calculate annual sales"

• "Example: Calculate annual sales by region"

• "Example: Calculate percentage of annual sales by region"

• "Example: Calculate city sales as a percentage of total sales"

Hint: You can also use analytic function templates to quickly and easily create calculations based on reporting functions (for more information, see "How to create a new calculation using an analytic function template".

Example: Calculate annual sales

This example calculates annual sales.

Figure A-18 Worksheet containing the Annual Sales calculation

Description of "Figure A-18 Worksheet containing the Annual Sales calculation"

The worksheet shows the annual sales value for cities in the year 2000.

Notes

• On table worksheets, the calculation displays the returned value for each row in the worksheet. To return a single value, move the calculation into the Page Items area (for more details see Additional Notes in "Example: Find the largest sales transactions in area with most sales trans'".

Example: Calculate annual sales by region

This example calculates the total annual sales by region.

.

Figure A-19 Worksheet containing the Annual Sales by Region calculation

Description of "Figure A-19 Worksheet containing the Annual Sales by Region calculation"

The worksheet shows the annual sales total for cities, grouped by region within year.

Example: Calculate percentage of annual sales by region

This example calculates the percentage of annual sales per region for each city in each year.

Figure A-20 Worksheet containing the % of Annual Sales calculation

Description of "Figure A-20 Worksheet containing the % of Annual Sales calculation"

The worksheet shows sales as a percentage of annual sales, grouped by region within year.

Example: Calculate city sales as a percentage of total sales

This example calculates city sales as a percentage of total sales.

Figure A-21 Worksheet containing the % of Annual Sales calculation

Description of "Figure A-21 Worksheet containing the % of Annual Sales calculation"

The worksheet shows the sales value for cities as a percentage of total sales.

Notes

• The function RATIO_TO_REPORT computes the ratio of a value to the sum of a set of values.

About LAG/LEAD functions

LAG and LEAD functions are typically used to compare values in different time periods. For example, compare sales figures in 2000 with sales figures in 2001.

• LAG - provides access to more than one row of a table at the same time without a self-join.

• LEAD - provides access to a row at a given offset after the current position.

You can also use windowing functions to compare values over time (see "Example: Compare sales figures across time using windowing".

Hint: You can also use analytic function templates to quickly and easily create calculations based on LAG/LEAD functions (for more information, see "How to create a new calculation using an analytic function template".

Examples:

• "Example: Compare sales figures across time using LAG/LEAD"

• "Example: Calculate sales growth across time"

• "Example: Rank sales growth"

Example: Compare sales figures across time using LAG/LEAD

In this example, you want to compare monthly sales figures with sales figures for the same month in the previous year. For example, to look at how January 1999 sales compare with January 1998 sales.

Figure A-22 Worksheet containing the Previous Year calculation partitioned by Calendar Month

Description of "Figure A-22 Worksheet containing the Previous Year calculation partitioned by Calendar Month"

The worksheet shows contains the calculation Previous Year, which shows for each Sales SUM amount the sales amount for one year previously. For example, the Previous Year value for January 1999 is $50889, which is the Sales SUM value for January 1998.

Notes

• Because there are no comparative figures for 1998, the Previous Year values for 1998 are blank.

• Notice that the value '1' in the LAG(Sales SUM,1) clause calculates the value from one year previously. For example, if you changed this value to '2', you would calculate the value from two years previously.

• Notice that the calculation includes the clause 'PARTITION BY Calendar Month', which gives you a value for each combination of Calendar Year (in the ORDER BY clause) and Calendar Month (in the PARTITION BY clause). In other words, the Previous Year value for February 1999 is the Sales SUM value for February 1998. If you removed this clause, you would calculate the value for the previous month (see example below). In other words, the Previous Year value for February 1999 would be the Sales SUM value for January 1999.

Figure A-23 Worksheet containing the Previous Year calculation with the partition removed

Description of "Figure A-23 Worksheet containing the Previous Year calculation with the partition removed"

Example: Calculate sales growth across time

In this example, you want to calculate the percentage growth of sales across years by comparing the sales figures with sales figures for the same month in the previous year. You will do this using the comparative sales figures from example "Example: Compare sales figures across time using LAG/LEAD".

Figure A-24 Worksheet containing the Growth calculation

Description of "Figure A-24 Worksheet containing the Growth calculation"

The worksheet shows contains the calculation Growth %, which shows for each month the percentage increase in sales since the previous year. For example, the Growth % value for January 1999 is 30.40% (i.e. from $50889 to $67887).

Notes

• Because there are no comparative figures for 1998, the Growth values for 1998 are blank.

• The calculation subtracts the Previous Year value from the Sales SUM value, then multiplies the result by the Sales SUM value divided by the Previous Year value. For example, if sales have risen from 75 to 100, the calculation becomes 25 * 1.33, giving 33.33% increase.

• For more information about the calculation Previous Year, see "Example: Compare sales figures across time using LAG/LEAD".

Example: Rank sales growth

In this example, you want to create a ranked list of sales growth, to show which months show the highest year on year increase in sales.

You will do this using the comparative sales figures and growth figures from examples "Example: Compare sales figures across time using LAG/LEAD" and "Example: Calculate sales growth across time", and a RANK function.

Figure A-25 Worksheet containing the Rank Growth calculation

Description of "Figure A-25 Worksheet containing the Rank Growth calculation"

The worksheet shows the ranked list position of sales growth. For example, the Rank Growth value for January 1999 is 3, which means that January was the third best performing month (i.e. the sales growth for the month of January between 1998 and 1999 was the third highest in the ranked list).

Notes

• Because there are no comparative figures for 1998, the Rank Growth values for 1998 are 1.

• Looking at the example above, you can see that sales growth was highest between October 1998 and October 1999 (40.02%)

• For more information about the calculation Previous Year, see "Example: Compare sales figures across time using LAG/LEAD".

• For more information about the calculation Growth, see "Example: Calculate sales growth across time".

About statistics functions

Statistics functions are used to compute covariance, correlation, and linear regression statistics. Each function operates on an unordered set. They also can be used as windowing and reporting functions.

Examples:

"Example: Calculate linear regression"

Example: Calculate linear regression

This example computes an ordinary least-squares regression line that expresses the Profit SUM per month as a linear function of its Sales SUM. The following functions are used:

• SLOPE - slope of determination of the regression line

• INTERCEPT - intercept of determination of the regression line

• REGR_R2 - coefficient of determination of the regression line

• REGR_COUNT - number of items

• REGR_AVGX - average sales

• REGR_AVGY - average profit

  Worksheet options	Set to:
  Items	Video Analysis Information: Calendar Year, Calendar Month, Sales SUM, Profit SUM
  Sort Order	Calendar Year
  Conditions	Department = Video Sale OR Department = Video Rental Calendar Year = 2000
  Calculations	Slope = REGR_SLOPE(Profit SUM,Sales SUM) OVER(PARTITION BY Calendar Year ORDER BY Profit SUM) Intercept = REGR_INTERCEPT(Profit SUM,Sales SUM) OVER(PARTITION BY Calendar Year ORDER BY Profit SUM) Coefficient = REGR_R2(Profit SUM,Sales SUM) OVER(PARTITION BY Calendar Year ORDER BY Profit SUM) Count = REGR_COUNT(Profit SUM,Sales SUM) OVER(PARTITION BY Calendar Year ORDER BY Profit SUM) Average = REGR_AVGX(Profit SUM,Sales SUM) OVER(PARTITION BY Calendar Year ORDER BY Profit SUM) Average 2 = REGR_AVGY(Profit SUM,Sales SUM) OVER(PARTITION BY Calendar Year ORDER BY Profit SUM)

  
  

Figure A-26 Worksheet containing the statistical calculations

Description of "Figure A-26 Worksheet containing the statistical calculations"

The worksheet shows for each month the slope, intercept, coefficient, count, and average values.

Notes

• For more information about regression analysis, refer to Oracle SQL Reference and Oracle Data Warehousing Guide.

Example of applying single conditions on analytic function items

In the first scenario, we apply two single conditions: Region = 'Central', and Rank <= 3 (where Rank is an analytic function).

• The Region = 'Central' condition is applied first, then Rank <= 3.

• Therefore, only sales figures for the Central region that have a ranking of three or less are included in the Results Set.

Example of applying multiple conditions on analytic function items

In the second scenario, we apply a multiple condition: Region = 'Central' AND Rank <= 3 (where Rank is an analytic function).

• The Rank <= 3 condition is applied first, then the Region = 'Central' condition.

• Therefore, only figures in the Central region that have an overall ranking of three or less are included in the Results Set.

Example: Compute the median profit using the PERCENTILE_DISC function

This example computes the median profit value for cities using the PERCENTILE_DISC function as a reporting aggregate function.

Figure A-27 Worksheet containing the Median(PERCENTILE_DISC) calculation

Description of "Figure A-27 Worksheet containing the Median(PERCENTILE_DISC) calculation"

The worksheet shows the median profit value for cities. The median profit value (i.e. 0.50 in the Cumulative Distribution column) is $61,942,21 (i.e. the value for Pittsburgh, which has the value 0.50 in the Cumulative Distribution column).

Notes

• On table worksheets, the calculation displays the returned value for each row in the worksheet. To return a single value, move the calculation into the Page Items area.

Example: Compute the median profit using the PERCENTILE_CONT function

This example computes the median profit value for cities using the PERCENTILE_CONT function as a reporting aggregate function.

Figure A-28 Worksheet containing the Median(PERCENTILE_CONT) calculation

Description of "Figure A-28 Worksheet containing the Median(PERCENTILE_CONT) calculation"

The worksheet shows the median profit value for cities. The median profit value is $63,076.41, which is the average profit value for the 0.50 and 0.55 percentile. In other words, the value for Pittsburgh plus the value for Denver, divided to two ($61,942.21 + $64,210.60)/2. For more information about how this function is calculated, see "About differences between PERCENTILE_CONT and PERCENTILE_DISC".

Notes

• If the number of rows in the calculation is even, the two middle results are averaged. In the example above, the values for the .50 and .55 percentile are averaged.

• On table worksheets, the calculation displays the returned value for each row in the worksheet. To return a single value, move the calculation into the Page Items area.

Example: Calculate hypothetical rank

This example calculates the hypothetical rank of profit values in relation to profit values for departments and regions. For example, to answer the question, how would a sales value of $500.00 be positioned in a ranked list of values for the Video Sale department in each region?

Note: This example uses a parameter to provide dynamic input to the calculation (for more information see "About using parameters to provide dynamic input to calculations").

Figure A-29 Worksheet containing the League table calculation

Description of "Figure A-29 Worksheet containing the League table calculation"

• The worksheet shows where the hypothetical value of $500 would rank in a ranked list of regions:

  • The hypothetical profit amount of $500 would rank number 2 in the Central region containing 520 profit values (range $13.49 to $537.78).

  • The hypothetical profit amount of $500 would rank number 7 in the East region containing 807 values (range $11.99 to $539.06).

    Notice that because the hypothetical amount ($500) is greater than the Profit MAX amount in the West region ($484.01), the amount $500 has the hypothetical rank of number 1 for the West region.

Notes

• The items Profit COUNT, Profit MAX, Profit MIN are not used in the calculation. They are displayed on the worksheet to help illustrate how the function is working. For example, if you can see that the Profit MAX value is $484.01 in the West region, you can see why a hypothetical value of $500.00 ranks as 1. This is because the hypothetical value is greater than the maximum value (i.e. the item Profit MAX).

• The League table calculation uses the value of the :Hypothetical Value parameter, entered when the worksheet is opened or refreshed. In the example below, the Hypothetical Value (displayed in the Hypothetical amount page item) is set to 500. For more information about using parameter values in calculations, see "About using parameters to provide dynamic input to calculations".

• The Rank function must take a non-aggregated value as an ORDER BY argument. For example, you could not perform this function on SUM(Profit) or Profit AVG.

• As an alternative to setting the Hypothetical Value as a parameter, you could enter the rank value directly into the calculation as the Rank() argument. For example:

RANK(500) WITHIN GROUP(ORDER BY Profit DESC NULLS FIRST)

• If you do not use a parameter, you will have to change the calculation to change the hypothetical value.

Example: Producing equi-width bands using WIDTH_BUCKET

This example divides profit figures into three bands according to their value.

Figure A-30 Worksheet containing the Equi-width bands calculation

Description of "Figure A-30 Worksheet containing the Equi-width bands calculation"

The worksheet shows equi-width bands for profit values for cities. The first band (0 to 9,999) contains Nashville, Minneapolis, Dallas, and Chicago. The second band (10,000 to 19,999) contains St. Louis. The third band (20,000 to 30,000) contains Cincinnati and Louisville.

Notes

• The WIDTH_BUCKET function takes four arguments:

  • worksheet item = Profit SUM

  • minimum value = 0

  • maximum value = 30000

  • number of bands = 3

• To assign bands in reverse order, reverse the minimum and maximum values. For example, WIDTH_BUCKET(Profit SUM,30000,0,3). This function produces the worksheet below.

Figure A-31 Worksheet containing the Equi-width bands calculation with reversed order

Description of "Figure A-31 Worksheet containing the Equi-width bands calculation with reversed order"

The worksheet shows equi-width bands for profit values for cities. The first band (20,000 to 30,000) contains Cincinnati and Louisville. The second band (10,000 to 19,999) contains St. Louis. The third band (0 to 9,999) contains Chicago, Dallas, Minneapolis, and Nashville.

Example: Find the largest sales transactions in area with most sales trans'

This example finds the largest sales transaction amount for the city with the most sales transactions in a period.

Figure A-32 Worksheet containing the Maximum sales in city with largest sales volume calculation

Description of "Figure A-32 Worksheet containing the Maximum sales in city with largest sales volume calculation"

The worksheet shows the largest sales transaction value in the city with the largest number of sales transactions. Cincinnati has the largest number of sales transactions (1220). The largest sales transaction for Cincinnati is $667.53.

Notes

• Sales MAX - contains the largest sales transaction amount.

• Sales COUNT - contains the number of sales transactions in the period.

• To apply the function, Discoverer does the following:

  • orders the Sales COUNT column in the database (default order is ascending)

  • takes the last value in Sales COUNT column (i.e. the LAST argument), which is the largest number, and looks up the city name for this row (Cincinnati)

  • orders transactions in the database for Cincinnati and returns the LAST value, $667.53 (the default order is ascending)

• The Sales COUNT and Sales MAX items are included to demonstrate that the calculation returns the correct result. The Sales COUNT and Sales MAX items are not used to calculate the result, which is calculated using aggregation in the database.

• Hint: On table worksheets, the calculation displays the returned value for each row in the worksheet. To return a single value, move the calculation into the Page Items area.

Example: Find the average sales transaction in area with least sales trans'

This example calculates the average sales transaction amount for the city with the smallest number of sales transactions in a period.

Figure A-33 Worksheet containing the Average sales in city with smallest sales volume calculation

Description of "Figure A-33 Worksheet containing the Average sales in city with smallest sales volume calculation"

The worksheet shows the average sales transaction value in the city with the smallest number of sales transactions. Nashville has the smallest number of transactions in the period (219). Therefore, the calculation returns the average transaction value for Nashville ($38.39).

Notes

• Sales COUNT - contains the number of sales transaction in the period.

• Sales AVG - contains the average sales transaction amount in the period.

• To apply the function, Discoverer does the following:

  • orders the Sales COUNT column in the database (default order is ascending)

  • takes the first value in Sales COUNT column (i.e. the FIRST argument), which is the smallest number, and looks up the city name for this row (Nashville)

  • calculates the average sales value for Nashville, $38.39

• The Sales COUNT and Sales AVG items are included to demonstrate that the calculation returns the correct result. The Sales COUNT and Sales AVG items are not used to calculate the result, which is calculated using aggregation in the database.

• Hint: On table worksheets, the calculation displays the returned value for each row in the worksheet. To return a single value, move the calculation into the Page Items area.