How to make Excel calculate if X is present within time period.

[ClausDaniel]

Hello,

First time posting, so let me just get straight to it.

I have a long period of time, where I have digital registration information about salary.

This includes Registration date for the payslip and number of working hours on the payslip.

So basically every month of salary has a line in Excel, and each line has registration date (column D) and number of hours (Column E). (and more).

Let's say I have 8 years of this data, and I need to know if there, anywhere in this 8 year period, is a 3 year period with 3848 or more hours. (The 3 year periode needs to be decided bu the registration date).

How would one do that? (keep in mind I am a novice in Excel).

Hopes my question makes sense.

Version is Microsoft office 365.

[#2]

@ClausDaniel 

=SUM(N(BYROW(B1:B19,LAMBDA(x,SUM(x:OFFSET(x,4,0))>60))))

I can't rebuild your dataset therefore i've set up a small example to check if the sum within 5 days exceeds 60. In this example only the sum from 08.01.2023 to 12.01.2023 exceeds 60.

[#3]

@OliverScheurich 

Thank you for your reply, maybe it is my lacking understanding of Excel in general, but I can't make your formula work for my dataset. 

Below is an example of a fictional dataset that would be like the ones I would have to do this task on.

So lets take the first row as an example, the registration date is 28-09-2023. Meaning that i need to know if there is 3848 in the column "hours", in the periode 29-09-2020 - 28-09-2023 (3 years). But that is just an example, like in your formula the period that has the 3848 hours can be in any 3 year period with in the allowed dataset. (It would also be preferable if the period that fulfills the 3848 hours, would be visible in some way. Maybe if Row A5 - A50 fulfills the 3848 hours, it would be shown that it is these specific rows that fulfill it).

So here is an example dataset:

Reg. date.	Hours
28-09-2023	29,5
30-08-2023	1,5
28-07-2023	29,75
28-06-2023	0
28-06-2023	68,54
29-06-2023	36,5
27-05-2023	0
27-05-2023	5,61
27-05-2023	28,75
27-04-2023	28,5
26-04-2023	68,58
26-04-2023	0
30-03-2023	6
29-03-2023	160,33
24-02-2023	152,83
25-02-2023	1,5
28-01-2023	7
27-01-2023	159,29
29-12-2022	20
28-12-2022	160,33
28-11-2022	160,33
29-11-2022	10
27-10-2022	160,33
28-10-2022	13,75
28-09-2022	152,83
29-08-2022	81,5
28-07-2022	10,75
28-07-2022	160,33
26-07-2022	41,25
29-06-2022	18,5
27-06-2022	96
26-05-2022	11,5
30-05-2022	99,75
28-04-2022	13,75
28-04-2022	120
30-03-2022	18,75
28-03-2022	88,75
25-02-2022	11,33
25-02-2022	55,25
10-01-2023	0
28-01-2022	61,25
29-12-2021	6,75
23-12-2021	130
27-11-2021	6,25
29-11-2021	84,75
28-10-2021	6,75
28-10-2021	132,25
29-09-2021	11,25
29-09-2021	102,5
26-04-2021	32,03
26-03-2021	84,35
24-02-2021	109,32
27-01-2021	93,38
27-01-2022	0
28-12-2020	81,09
22-12-2020	39
23-11-2020	76
26-10-2020	60
24-09-2020	60
24-08-2020	83,25
23-07-2020	62
28-03-2020	32,75
27-02-2020	79,5
19-01-2021	0
28-11-2019	160,33
27-11-2019	160,33
25-10-2019	160,33
28-09-2019	50,27
25-08-2019	27
27-07-2019	30,56
27-06-2019	82,02
29-05-2019	100

[#4]

@ClausDaniel 

=LET(timeperiod,BYROW(SEQUENCE(A2-A73,,0),LAMBDA(x,IF(SUM(FILTER(B2:B73,(A2:A73<=DATE(YEAR(A2),MONTH(A2),DAY(A2)-x))*(A2:A73>=DATE(YEAR(A2)-3,MONTH(A2),DAY(A2)-x+1))))>3550,TEXT(DATE(YEAR(A2),MONTH(A2),DAY(A2)-x),"DD.MM.YYYY")&"-"&TEXT(DATE(YEAR(A2)-3,MONTH(A2),DAY(A2)-x+1),"DD.MM.YYYY"),""))),

len,BYROW(timeperiod,LAMBDA(x,LEN(x))),

IFERROR(FILTER(timeperiod,len),"no result"))

This formula returns the periods for which the sum within 3 years is above 3550. In the sample data the formula would return "no result" for 3848 because there isn't a 3 year period with sum over 3848.

The assumption is that 28.09.2023 is the latest date and 29.05.2019 is the earliest date in the Reg. date. column. Then the formula calculates the sum for all ranges

28.09.2023	29.09.2020
27.09.2023	28.09.2020
26.09.2023	27.09.2020
25.09.2023	26.09.2020
24.09.2023	25.09.2020

and so on. Then it checks for each range if the sum is above 3550 and if so it returns the time range otherwise a blank cell. In the last step the data is filtered to only return the ranges with sum above 3550. For the sample data there are 24 time ranges of 3 years which sum up to over 3550. The sample data is in range A2:B73.

I'd sort the data in descending order of the Reg. date. column then it would be possible to return a result such as e.g. A5:A50 if the sum in this range is greater than 3550.

[#5]

This seems like just the thing @OliverScheurich
I was getting errors on "FILTER" though, so I changed it to the danish "FILTRER". Which means "to filter".

After that I'm getting #NAVN? (#NAME?) error. From a quick google search that indicates an error in the formula name, but isn't the formula name here "LET"? (Again excuse my lack of knowledge).

Here is the formula I have at this moment:

=LET(timeperiod;BYROW(SEQUENCE(A2-A73;;0);LAMBDA(x;IF(SUM(FILTRER(B2:B73;(A2:A73<=DATE(YEAR(A2);MONTH(A2);DAY(A2)-x))*(A2:A73>=DATE(YEAR(A2)-3;MONTH(A2);DAY(A2)-x+1))))>3550;TEXT(DATE(YEAR(A2);MONTH(A2);DAY(A2)-x);"DD.MM.YYYY")&"-"&TEXT(DATE(YEAR(A2)-3;MONTH(A2);DAY(A2)-x+1);"DD.MM.YYYY");"")));len;BYROW(timeperiod;LAMBDA(x;LEN(x)));IFERROR(FILTRER(timeperiod;len);"no result"))

Thank you for your time and effort.

[#6 · Accepted solution]

@ClausDaniel 

If you work with danish Excel the whole formula has to be translated into danish. If you open my attached file the formula is translated according to your regional setting. The formula only works in Excel for Microsoft 365, Excel for Microsoft 365 for Mac and Excel for the web.

Below is what i get from a translator but i assume that in danish Excel there isn't a BYROW formula and the TEKST format "DD.MM.YYYY" is certainly different in danish Excel. And i wonder if SUM, LAMBDA and LET are the same in english and danish Excel.

=LET(timeperiod;BYROW(SEKVENS(A2-A73;;0);LAMBDA(x;HVIS(SUM(FILTRER(B2:B73;(A2:A73<=DATO(ÅR(A2);MÅNED(A2);DAG(A2)-x))*(A2:A73>=DATO(ÅR(A2)-3;MÅNED(A2);DAG(A2)-x+1))))>3550;TEKST(DATO(ÅR(A2);MÅNED(A2);DAG(A2)-x);"DD.MM.YYYY")&"-"&TEKST(DATO(ÅR(A2)-3;MÅNED(A2);DAG(A2)-x+1);"DD.MM.YYYY");"")));

len;BYROW(timeperiod;LAMBDA(x;LÆNGDE(x)));

HVIS.FEJL(FILTRER(timeperiod;len);"no result"))

[#7]

@OliverScheurich thank you so much. :)

Changing the date format to DD.MM.ÅÅÅÅ fixed it, seems as though LAMBDA and LET are indeed the same in Danish, SUM is also i just wasn't uncertain about that one.

Again thank you for your time and effort, it is very much appreciated.

[#8]

@ClausDaniel 

You are welcome. Thank you for your very interesting question.