I would like a column displaying the fiscal week. Our fiscal year begins in April.
So far I have the below, using datename(ww,DateAndTime) as Week
DateAndTime Week
2015-04-01 22:45 14
2015-06-14 13:22 25
2015-12-02 09:15 49
2016-01-01 07:35 1
I would like the output to show:
DateAndTime Week Fiscal Week
2015-04-01 22:45 14 1
2015-06-14 13:22 25 12
2015-12-02 09:15 49 36
2016-01-01 07:35 1 41
While I don't understand the logic behind the fiscal week (the difference between 1 and 41 is 40, but between 14 and 1 it's 39), maybe I'm missing something or you made a typo.
However, in general you'd do something like this (assuming the difference is 40 weeks):
SELECT week, (week+40)%52 AS fw FROM ...
If the fiscal year starts at a different week every each (say, 13th or 14th week depending on year), you can use the date and time functions, but they may vary between SQL versions. In MySQL you have YEAR(), MONTH(), WEEK(), etc.
For example:
SELECT week, (week+(52-WEEK(CONCAT_WS('-', YEAR(NOW()), '04-01'))))%52 FROM ...
But it might be overkill.
Note: It is possible to count the other way: if you subtract the diff from the week instead of adding, you will need to add 52 if the number is negative. You can do that by adding 52 and then doing modulo (%) 52.
Related
I have a sample of a table as below:
Customer Ref
Bear Rate
Distance
Month
Revenue
ABA-IFNL-001
1000
01/01/2022
-135
ABA-IFNL-001
1000
01/02/2022
-135
ABA-IFNL-001
1000
01/03/2022
-135
ABA-IFNL-001
1000
01/04/2022
-135
ABA-IFNL-001
1000
01/05/2022
-135
ABA-IFNL-001
1000
01/06/2022
-135
I also have a sample of a calendar table as below:
Date
Year
Week
Quarter
WeekDay
Qtr Start
Qtr End
Week Day
04/11/2022
2022
45
4
Fri
30/09/2022
29/12/2022
1
05/11/2022
2022
45
4
Sat
30/09/2022
29/12/2022
2
06/11/2022
2022
45
4
Sun
30/09/2022
29/12/2022
3
07/11/2022
2022
45
4
Mon
30/09/2022
29/12/2022
4
08/11/2022
2022
45
4
Tue
30/09/2022
29/12/2022
5
09/11/2022
2022
45
4
Wed
30/09/2022
29/12/2022
6
10/11/2022
2022
45
4
Thu
30/09/2022
29/12/2022
7
11/11/2022
2022
46
4
Fri
30/09/2022
29/12/2022
1
12/11/2022
2022
46
4
Sat
30/09/2022
29/12/2022
2
13/11/2022
2022
46
4
Sun
30/09/2022
29/12/2022
3
14/11/2022
2022
46
4
Mon
30/09/2022
29/12/2022
4
15/11/2022
2022
46
4
Tue
30/09/2022
29/12/2022
5
16/11/2022
2022
46
4
Wed
30/09/2022
29/12/2022
6
17/11/2022
2022
46
4
Thu
30/09/2022
29/12/2022
7
How can I join/link the tables to report on revenue over weekly and quarterly periods using the calendar table? I can put into two tables if needed as an output eg:
Quarter Starting
31/12/2021
01/04/2022
01/07/2022
30/09/2022
Quarter
1
2
3
4
Revenue
500
400
540
540
Week Date Start
31/12/2021
07/01/2022
14/01/2022
21/01/2022
Week
41
42
43
44
Revenue
33.75
33.75
33.75
33.75
I am using alteryx for this but wouldnt mind explaination of possible logic in sql to apply it into the system
Thanks
Before I get into the answer, you're going to have an issue regarding data integrity. All the revenue data is aggregated at a monthly level, where your quarters start and end on someday within the month.
For example - Q4 starts September 30th (Friday) and ends Dec. 29th (Thursday). You may have a day or two that bleeds from another month into the quarters which might throw off the data a bit (esp. if there's a large amount of revenue during the days that bleed into a quarter.
Additionally, your revenue is aggregated at a monthly level - unless you have more granular data (weekly, daily would be best), it doesn't make sense to do a weekly calculation since you'll probably just be dividing revenue by 4.
That being said - You'll want to use a cross tab feature in alteryx to get the data how you want it. But before you do that, we want to aggregate your data at a quarterly level first.
You can do this with an if statement or some other data cleansing tool (sorry, been a while since I used alteryx). Something like:
# Pseudo code - this won't actually work!
# For determining quarter
if (month) between (30/09/2022,29/12/2022) then 4
where you can derive the logic from your calendar table. Then once you have the quarter, you can join in the Quarter Start date based on your quarter calculation.
Now you have a nice clean table that might look something like this:
Month
Revenue
Quarter
Quarter Start Date
01/01/2022
-135
4
30/09/2022
01/01/2022
-135
4
30/09/2022
Aggregate on your quarter to get a cleaner table
Quarter Start Date
Quarter
revenue
30/09/2022
4
300
Then use cross tab, where you pivot on the Quarter start date.
For SQL, you'd be pivoting the data. Essentially, taking the value from a row of data, and converting it into a column. It will look a bit janky because the data is so customized, but here's a good question that goes over pivioting - Simple way to transpose columns and rows in SQL?
i have a hard time figuring out how to compare the same period (e.g. iso week 48) from different years for a certain metric in different columns. I am new to SQL and haven't fully understand how PARTITION BY works but guess that i'll need it for my desired output.
How can i sum the data from column "metric" and compare same periods of different date ranges (e.g. YEAR) in a table?
current table
date iso_week iso_year metric
2021-12-01 48 2021 1000
2021-11-30 48 2021 850
...
2020-11-28 48 2020 800
2020-11-27 48 2020 950
...
2019-11-27 48 2019 700
2019-11-26 48 2019 820
desired output
iso_week metric_thisYear metric_prevYear metric_prev2Year
48 1850 1750 1520
...
Consider below simple approach
select * from (
select * except(date)
from your_table
)
pivot (sum(metric) as metric for iso_year in (2021, 2020, 2019))
if applied to sample data in your question - output is
I have a table in snowflake that looks like this:
starting_date fiscal_week calendar_week
2020-12-31 53 53
2021-01-01 53 53
2021-01-08 53 01
The fiscal_week and calendar_week look up off of the starting_date (format is YYYYMMDD). The fiscal week should follow the 1-53 week number in the year. The calendar_week does not necessarily need to start off on a Sunday, it just needs to start off on the first day of the year as week 01.
Thus, I would like it to look like this:
starting_date fiscal_week calendar_week
2020-12-31 53 53
2021-01-01 53 01
2021-01-08 53 02
I've tried using the weekiso and week functions such as below but still am not getting the desired results. The results still reflect what is shown in the first table.
select *, weekiso (starting_date) as calendar_week from my table
Is there a specific way to do this in the snowflake database? I'm assuming that's where my issue is coming from but I'm not sure.
I'm new to using snowflake so any help would be appreciated!
You can extract days of the year and do some arithmetic on that. It's probably easier to just change the week of year policy at session level though
select floor((dayofyear(starting_date)+6)/7)
The below gives me week numbers where week 1 starts on 1/4/2021
date_trunc('week', transaction_date) as week_number
How can I create a week_number where the week starts on January 1st and counts up 7 days for every week thereafter (for every year)?
And round up/down to 52 weeks at the end of the year?
Code attempted:
This doesn't give me the answer, but I'm thinking something like this might work...
ceil(extract(day from transaction_date)/7) as week_number
Expected Output:
transaction_date
week_number
1/1/2020
1
1/8/2020
2
...
...
12/31/2020
52
1/1/2021
1
1/8/2021
2
...
...
12/27/2021
52
12/28/2021
52
12/29/2021
52
12/30/2021
52
12/31/2021
52
1/1/2022
1
Thanks in advance!
A simple way is to use date arithmetic:
select 1 + (transaction_date - date_trunc('year', transaction_date)) / 7 as year_week
The below gives me week numbers where week 1 starts on 1/4/2021
It is the default behaviour and it is defined that way in ISO.
WEEK_OF_YEAR_POLICY
Type Session — Can be set for Account » User » Session
Description
Specifies how the weeks in a given year are computed.
Values
0: The semantics used are equivalent to the ISO semantics, in which a week belongs to a given year if at least 4 days of that week are in that year.
1: January 1 is included in the first week of the year and December 31 is included in the last week of the year.
Default 0 (i.e. ISO-like behavior)
It could be overrriden on multiple levels. The most granular is on the session level:
ALTER SESSION SET WEEK_OF_YEAR_POLICY = 1;
Then you could use the standard code:
SELECT date_trunc('week', transaction_date) as week_number
FROM ...;
So here is my scenario.
If it's 20th week of the Year 2016, I need to know it's 5/8/2016. This is consistent for the rest of week. Same goes for all the week.
Week# WeekDate
29 7/10/2016 0:00
30 7/17/2016 0:00
28 7/3/2016 0:00
31 7/24/2016 0:00
32 7/31/2016 0:00
33 8/7/2016 0:00
34 8/14/2016 0:00
dateadd(DAY,-(datepart(weekday,DATEADD(WEEK,30-1,CAST('2016'||'-1-1' AS DATE)))-1),DATEADD(WEEK,30-1,CAST('2016'||'-1-1' AS DATE)))
This works with Intersystems. You will have to construct your own date with the known year value and replace '30' with the known week value from your table
Thanks for the contribution, I did something like below and it worked. This is keeping Sunday as the first day of week.
If you want to use Monday or Tuesday, add +2, +3 in the WeekNum calculation below.
DATE(SampleTableWithDate.Year],1,1) - WEEKDAY(DATE(SampleTableWithDate.Year,1,1),1) + (WEEKNUM(SampleTableWithDate.Date)-1)*7 + 1
Find the First Date of this Year - First Day of Week This Year + Add Weeks till Date