Forum Discussion
Pivot Table group positive and negative data
Attached an example with a single [Amount] column with Positive & Negative values:
Edit the query in the attachment and follow the steps in APPLIED STEPS (all done with the UI). Once you've separated your [Income] & [Expense] (the added Custom steps) and removed [Amount] that's no longer necessary, right-click on column [Category] and in the drop-down menu select Unpivot Other Columns
IMPORTANT:
- For [Expense] you'll see I used function Number.Abs to convert to negative [Amount]s to positive. If you want your Expense to show negative in your PivotTable then remove Number.Abs (you will have to adjust the Graph DAX measure)
- When you add Custom Columns [Income] & [Expense] Power Query doesn't auto. Type the columns. Taking ex. [Income] when you create it, it displays exactly as below in PQ formula bar:
You must edit the formula and add, before ): , type number then Validate:
You must do the same for the [Expense] column
Reason is: default data Type in Power Query is Type any. Column(s) of Type any loaded to the Excel Data Model will be assigned the Type Text (no Sum, Avg... possible)
Thanks for your quick response. Could you please let me know how you carried out unpivoting step which resulted in categorizing the data as rows and columns.
Also, my raw data has a single column of income and expenses, income in positive values and expense as negative values. Does this mean i need to split them into separate column, positive and negative and then carryout your method? Appreciate your support on this.
Attached an example with a single [Amount] column with Positive & Negative values:
Edit the query in the attachment and follow the steps in APPLIED STEPS (all done with the UI). Once you've separated your [Income] & [Expense] (the added Custom steps) and removed [Amount] that's no longer necessary, right-click on column [Category] and in the drop-down menu select Unpivot Other Columns
IMPORTANT:
- For [Expense] you'll see I used function Number.Abs to convert to negative [Amount]s to positive. If you want your Expense to show negative in your PivotTable then remove Number.Abs (you will have to adjust the Graph DAX measure)
- When you add Custom Columns [Income] & [Expense] Power Query doesn't auto. Type the columns. Taking ex. [Income] when you create it, it displays exactly as below in PQ formula bar:
You must edit the formula and add, before ): , type number then Validate:
You must do the same for the [Expense] column
Reason is: default data Type in Power Query is Type any. Column(s) of Type any loaded to the Excel Data Model will be assigned the Type Text (no Sum, Avg... possible)
There is a problem with this approach. As you can see, with this approach, as transfers has both positive and negative values in the data, it gets listed in both income and expense. With this approach I'm unable to calculate my net income and net expenses. Could you please help with this?
- can you upload your data and expected result?
Note quite sure this will do exactly what you expect....: Aggregate the amounts for the Transfers category. Then, if the aggregated amount is >= 0 Transfers will be reported as Income otherwise as Expense
let Source = Excel.CurrentWorkbook(){[Name="SourceTable"]}[Content], ChangedTypes = Table.TransformColumnTypes(Source,{{"Category", type text}, {"Amount", type number}}), Transfers = Table.Group( Table.SelectRows(ChangedTypes, each ([Category] = "Transfers")), {"Category"}, {{"Amount", each List.Sum([Amount]), type nullable number}} ), OtherCategories = Table.SelectRows(ChangedTypes, each ([Category] <> "Transfers")), Combined = Table.Combine({OtherCategories, Transfers}), AddedIncome = Table.AddColumn(Combined, "Income", each if [Amount] >= 0 then [Amount] else null, type number ), AddedExpense = Table.AddColumn(AddedIncome, "Expense", each if [Amount] < 0 then Number.Abs([Amount]) else null, type number ), RemovedAmount = Table.RemoveColumns(AddedExpense,{"Amount"}), UnpivotedColumns = Table.UnpivotOtherColumns(RemovedAmount, {"Category"}, "Attribute", "Value") in UnpivotedColumns
