question

Upvotes
Accepted
marlyndb avatar image
3 ●0 ●0 ●3

Data from 2016 changes depending on when I retrieve it

I retrieved data from 2016 until 2020 around 2 weeks ago. Today I retrieved the exact same data but I got different results. This was for example for market value, total assets, beta coefficient,.... How is it possible that data from 2016 still changes now? In the image you can see the data of the beta coefficient. The bold, underlined data is retrieved 2 weeks ago, the other data is from today. You can see a clear difference between both data, while it is supposed to be the exact same. How is this possible?


2016: 0,742    2016: 0,692
      0,751          0,667
      1,21           1,217
      1,2            1,22
      1,2495         1,2404
2017: 0,841    2017: 0,754
      0,995          0,991
      1,085          1,097
      1,31           1,21
      0,8387         0,9101


datastream
icon clock
10 |1500

Up to 2 attachments (including images) can be used with a maximum of 512.0 KiB each and 1.0 MiB total.

raksina.samasiri avatar image   raksina.samasiri

hi @marlyndb ,

Thank you for your participation in the forum. Is the reply below satisfactory in resolving your query?
If so please can you click the 'Accept' text next to the appropriate reply? This will guide all community members who have a similar question.

Thanks,
AHS

3 Answers

· Write an Answer
Upvotes
Accepted
Jirapongse avatar image
79.1k ●250 ●52 ●74

@marlyndb

I checked and found the Datastream REST API may not support only a year format (2016, 2022) in the start and end parameters. The start and end should be relative dates or absolute dates.

I checked with Excel with the following formula.

=@DSGRID("U:IBM","DWTA","2016","2022","Y","RowHeader=true;ColHeader=true;DispSeriesDescription=false;YearlyTSFormat=false;QuarterlyTSFormat=false","")

It returns:

NameINTERNATIONAL BUS.MCHS. - TOTAL ASSETS2016105673000201711224600020181204940002019118166000202014700400020211467300002022124631000


Next, I changed YearlyTSFormat to true.

=@DSGRID("U:IBM","DWTA","2016","2022","Y","RowHeader=true;ColHeader=true;DispSeriesDescription=false;YearlyTSFormat=true;QuarterlyTSFormat=false","")

The output is:

NameINTERNATIONAL BUS.MCHS. - TOTAL ASSETS3/28/20161056730003/28/20171122460003/28/20181204940003/28/20191181660003/28/20201470040003/28/20211467300003/28/2022124631000


I think that the values are for absolute dates but the dates were truncated to years in the first formula.

However, you may contact Datastream Product Support directly to confirm it.

icon clock
10 |1500

Up to 2 attachments (including images) can be used with a maximum of 512.0 KiB each and 1.0 MiB total.

Upvotes
zoya faberov avatar image
32.2k ●40 ●11 ●20

Hello @marlyndb ,

What you describe appears to be a suspected content issue within product, rather then API usage question or issue.

If you would like to include a code segment to reproduce the results displayed, we can verify the issue, by running the same code on our side, twice, with the interval of two weeks time, and see if the results differ.

You may also report the issue directly to Datastream Product Support for investigation. Please include the complete details including the support request then.

icon clock
10 |1500

Up to 2 attachments (including images) can be used with a maximum of 512.0 KiB each and 1.0 MiB total.

Upvotes
Jirapongse avatar image
79.1k ●250 ●52 ●74

@marlyndb

I assume that you are using relative dates so data may belong to different dates.

For example, the following codes provide different results.

ds.get_data(tickers="897E(TW:TSM)",  start = "2010-01-01", end = "2020-01-01", freq = "Y")

2010-01-01 0.68

2011-01-01 0.68

2012-01-01 0.62

ds.get_data(tickers="897E(TW:TSM)",  start = "2010-01-08", end = "2020-01-01", freq = "Y")

2010-01-08 0.78

2011-01-08 0.76

2012-01-08 0.68

ds.get_data(tickers="897E(TW:TSM)",  start = "-12Y", freq = "Y")

2010-03-28 0.83

2011-03-28 0.82

2012-03-28 0.73

You may try to use adsolute dates instead.


icon clock
10 |1500

Up to 2 attachments (including images) can be used with a maximum of 512.0 KiB each and 1.0 MiB total.

marlyndb avatar image marlyndb

@Jirapongse Is it possible that I did the retrieval wrong? I always used the year (yyyy) itself as a start and end date instead of the yyyy-mm-dd format. Is it possible that the database then just chooses which exact date it retrieves?

So I did:

ds.get_data(tickers="897E(TW:TSM)", start = "2016", end = "2020", freq = "Y")

Instead of

ds.get_data(tickers="897E(TW:TSM)", start = "2016-01-01", end = "2020-01-01", freq = "Y")

Can that be my mistake? (I retrieve the data via excel)

Jirapongse avatar image Jirapongse marlyndb

@marlyndb

I am unable to run the code below.

ds.get_data(tickers="897E(TW:TSM)", start = "2016", end = "2020", freq = "Y")

It returns:

$$ER: E105,INVALID START DATE ENTERED  
marlyndb avatar image marlyndb Jirapongse

@Jirapongse

For example to get the total assets of a firm I used this formula in excel:

=@Thomson.Reuters.AFOSpreadsheetFormulas.DSGRID(Sheet1!$B$1:$B$80;"DWTA";"2016";"2020";"Y";"Sym=RIC";"")

And I get different results depending on the day that I use it.

Write an Answer

Hint: Notify or tag a user in this post by typing @username.

Up to 2 attachments (including images) can be used with a maximum of 512.0 KiB each and 1.0 MiB total.