Is there a way to count the number of distinct values in a filtered column in Excel?
Using the formula at https://exceljet.net/formula/count-unique-values-in-a-range-with-countif I can count the number of distinct values in a range, but when that range has been filtered with an auto-filter I still get the unfiltered count.
For example:
A B
1 Scarf Blue
2 Hat Red
3 Gloves Green
4 Coat Blue
5 Balloon Red
6 Shoes Blue
Counting unique values in B with =SUMPRODUCT((B1:B6<>"") / COUNTIF(B1:B6,B1:B6 & "")) should return 3 as the distinct values are Red, Green and Blue.
If I auto filter Column B to just select Red items, the resulting table will look like:
A B
2 Hat Red
5 Balloon Red
In this case the number of distinct values retuned should be 1. But the formula above still returns 3.
The formula should also cope with multiple selections in the auto-filter, so for example filtering for Blue and Green should result in the following table:
A B
1 Scarf Blue
3 Gloves Green
4 Coat Blue
6 Shoes Blue
From which the formula should return 2 (Blue, Green).
Finally, if I am filtering on column A rather than B, the formula should still work. So If I am only interested in Hat, Scarf and Coat, filtering column A for these values would result in:
A B
1 Scarf Blue
2 Hat Red
4 Coat Blue
From which the formula should return 2.
(I'm using Excel 2013 and need to do this in a formula rather than using VBA etc)
I also found this page on office.com which I thought might help, but alas I can't get it to work for me.
This reference shows how you can exclude hidden rows using AGGREGATE
Excluding hidden rows with AGGREGATE
You can then use a standard way of counting unique values like this
Counting unique values with FREQUENCY
So if you were counting values in column B, you would need a helper column (say C) containing
=IF(AGGREGATE(3,5,B2),B2,"")
Then plug in the form of count unique that ignores empty cells
=SUM(IF(FREQUENCY(IF(LEN(C2:C10)>0,MATCH(C2:C10,C2:C10,0),""), IF(LEN(C2:C10)>0,MATCH(C2:C10,C2:C10,0),""))>0,1))
Or your formula if you prefer
=SUMPRODUCT((C2:C10<>"") / COUNTIF(C2:C10,C2:C10 & ""))
Related
I have 2 list. Both have a lot of column. I would like to insert a column from Sheet2 to Sheet1 based on a ceratin key column. Also sheet2 have much more rows than sheet1 so it ll be inserted only partly and still there ll be elements with no matches. For an example:
Sheet1:
Names ID Car Color
John 1 Audi Empty
Andy 4 Toyota Empty
Mike 3 BMW Empty
Tony 2 Suzuki Empty
Sheet2:
ID Cost Color
6 200 Blue
3 200 Red
4 300 Green
5 100 Red
1 50 Black
I would like to get the "color" from Sheet2 to Sheet1 by using the "ID". Using Excel 2010. I suspect I need INDEX+MATCH combination, but the examples I can find are not detailed and more simple so I coudn't figure out how to use them.
How about inserting this formula on Column D in the first row, then dropping the formula down:
=VLOOKUP(B1,Sheet2!$1:$1048576,3,FALSE)
Or to find the column that contains "Color", use Index Match Match, as follows:
=INDEX(Sheet2!$1:$1048576,MATCH(Sheet1!B2,Sheet2!A:A,0),MATCH("Color",Sheet2!$1:$1,0))
This will find the value in Column B in Sheet2 and give you the row number, then it will find the Column that contains "Color" and return the Column number, with those two numbers Index will return your color.
I have 8 combinations o triplets, each with a name attached to them. For example:
Category a = 1,2,3
Category b = 3,1,2
etc...
Each value in the triplet is either 1, 2 or 3.
I need to be able to input 3 values (also 1,2 or 3) into 3 cells, and I need the corresponding category to be displayed next to them (based on the 8 hard coded variants).
For example: if I were to enter 3 1 2 into the three cells, "Category b" would show up in the 4th cell.
Here is a screenshot of what I need:
The yellow cells are hard coded.
I have a long list of green cells.
I need to calculate the category in the blue cells based on the green cells.
use this in E2:
=INDEX(G:G,AGGREGATE(15,6,ROW($G$2:$G$9)/(($H$2:$H$9=B2)*($I$2:$I$9=C2)*($J$2:$J$9=D2)),1))
Then copy down
I have a data like this in excel:
1 1
1 2
2 2
3 2
And I want the result to be displayed like this:
1
1
1
2
2
2
2
3
Its like a ascending sorting. If the number in columnA <= columnB, it will be display first, at the same time, columnB will have a empty cell or blank value.
I have looked for this for a while but no one has the problem like this.
I would prefer a formula rather than VBA code.
Please advise, thanks!
Edit: I hope this explanation will be easier to understand
Data (A1:B2) - Result (C1:D4). Each number in Data (A1:B2) is a result of complicated formula so I cannot change its position
A B C D
1 1 2 1
2 3 4 2
3 3
4 4
Honestly, if you want to avoid VBA, here's a manual process that would give you the results you want:
Color each column (ie. Column A Red and Column B Blue)
Place values of column B under column A
Sort & Filter by Sort Smallest to Largest
Copy the contents of column A to column B
Last for Column A Find and Replace all* with format Blue then do the same for column B only with format Red. (See below for more info on how to do that).
Following those steps isn't the most conventional, but it will give you the results you wanted without VBA.
*Find and Replace based on font color:
1. Press Ctrl-H
2. Click Options
3. Click Format > Font > then select your font color > OK
4. With the Find What blank with the font color and the Replace With blank click Replace All.
I have a sheet of data with multiple dimensions like this:
A B C D E
1 COUNTRY FLAVOUR SIZE DATE SALES ($)
2 Japan Strawberry 100ml 10/12/14 100
3 Japan Banana 100ml 10/03/15 100
4 China Orange 200ml 14/04/15 30
5 France Strawberry 200ml 11/04/15 400
6 UK 200ml 23/04/15 250
7 ....
I want to aggregate this data over a date range, where the summary sheet has each dimension (country & flavour), and if I do not specify a dimension it sums all rows for that dimension.
A B C
1 COUNTRY FLAVOUR SALES TOTAL
2 Japan Strawberry 100
3 Japan 200
4 Strawberry 500
I can do this if all the dimensions are present (i.e. row 2 above) using a SUMPRODUCT or SUMIFS:
=SUMPRODUCT((data!A$2:A$100=A1)*(data!B$2:B$100=B1)*(data!D$2:D$100>[start_date]*(data!D$2:D$100<[end_date])*(data!E$2:E$100))
However I have not been able to figure out how to include all rows for a dimension if that input cell is empty (e.g. row 3 above). I tried:
Adding an IF statement or OR statement within the criteria (e.g. OR(data!A$2:A$100=A1,isblank(A1))).
Using a + in a SUMPRODUCT as an OR statement, (per this answer https://stackoverflow.com/a/27536131/1450420)
One solution is to have different branches of the formula depending on which summary dimensions are present, but that would quickly get out of control if I extend this same behaviour to further dimensions like Size.
Any help appreciated!
(I'm running Excel Mac 2011).
EDIT
Per #BrakNicku's comment one of the formulas I tried was =SUMPRODUCT(((data!A$2:A$100=A2)+ISBLANK(A2))*((data!B$2:B$100=B2)+ISBLANK(B2))*(data!E$2:E$100))
The reason this doesn't work is that sometimes my data has blank attributes (edited above). For some reason this formula double-counts rows where the attribute present matches (e.g. data!A6) but the other attribute is missing (e.g. data!B6).
EDIT 2
I can see why this double-counting is happening, because the + is summing the match because data!A$2:A$100=A2 (they match because they are both blank) and the match because ISBLANK(A2) (it is indeed blank). The question would remain how to achieve this without double counting. If needed a workaround could be to fill all blank cells on my data with some placeholder value.
The reason for double-counting values is here:
((data!A$2:A$100=A2)+ISBLANK(A2))
If a cell in A column is blank, both parts of this sum are equal 1. To get rid of this problem you can change it to:
(((data!A$2:A$100=A2)+ISBLANK(A2))>0)
Try this (I only included the first two, I left the dates out):
=SUMPRODUCT((((Data!$A$2:$A$5=A2)+(A2=""))>0)*(((Data!$B$2:$B$5=B2)+(B2=""))>0)*(Data!$E$2:$E$5))
I'm trying to get a count of matches from two columns. Let's say I have a table with the following data.
Cars
In column A I have colours of cars e.g. Red, Yellow, Black, Green
In column B I have the type of car e.g. Diesel, Automatic, Manual
I want a count of all red cards that run on diesel so which function would I use?
Column A has 3 red cars
Column B has 6 diesel cars in total but only 1 of the red cars is a diesel.
I've tried :
=COUNTIF(A2:A10,"Red")+COUNTIF(B2:B10,"Diesel")
But it gives a combined count of all red cars and all diesel cars (9), not all red diesel cars, which should be 1.
You could use the =COUNTIFS() formula:
=COUNTIFS(A1:A10,"Red",B1:B10,"Diesel")
You choud do an operation for each row, and sum all.