Serving the Quantitative Finance Community

 
User avatar
gevor
Topic Author
Posts: 0
Joined: July 20th, 2006, 9:10 pm

Analyst data

July 6th, 2007, 7:32 am

Hi,The more I look at the analyst historical data, published py different agencies, the more concerns about the quality of that data I have.So guys, what is the best data for analyst estimates?
 
User avatar
csa
Posts: 0
Joined: February 21st, 2003, 3:16 am

Analyst data

July 8th, 2007, 8:58 pm

Are you talking about EPS estimates? IBES and First Calls.
 
User avatar
gevor
Topic Author
Posts: 0
Joined: July 20th, 2006, 9:10 pm

Analyst data

July 9th, 2007, 7:50 am

Hm..I thought IBES has some problems with data accuracy and backfills, which limit it's use for historical backtests.And what do you think about Reuters consensus estimates?
 
User avatar
csa
Posts: 0
Joined: February 21st, 2003, 3:16 am

Analyst data

July 10th, 2007, 3:34 am

If I recall correctly, IBES is now part of Thomson Financial and their data should be the same as First Calls. The adjustments that IBES or First Calls make are necessary to make the estimates by different analysts comparable. Also, adjustments are made to try and make things much more consistent (I believe they account for stock splits) but there are indeed a lot of errors. There is probably no better source though, which is why they can get away with it. My understanding of First Calls is that analysts upload their data to the site, which gives it a date and time stamp. You can actually get the different estimates by analysts with date and time stamps but you have to clean the data to get what you want. IBES, on the other hand, is an estimate made "somewhere in the middle of the month" - whatever that means. So, if you wanted to make assumptions, there is a wide range of possible dates where the previous IBES estimate could still hold and it is not the same every month.As for Reuters, their service is called Multex. Multex provides a similar service to First Call, but is not as popular. The consensus estimates are also done in a similar fashion. As for accuracy, I think they are all the same - there's not much we can do about it though.
 
User avatar
gevor
Topic Author
Posts: 0
Joined: July 20th, 2006, 9:10 pm

Analyst data

July 10th, 2007, 7:26 am

Sad :-(You are right, Reuters is not that popular, sometimes you have 1-2 analysts for Reuters and a few more for IBES.So, should I go for IBES, do backtests, run the strategy live and just hope that the data was not too screwed?..
 
User avatar
Wibble
Posts: 1
Joined: January 23rd, 2004, 3:15 pm

Analyst data

July 10th, 2007, 10:24 am

our analysts pay for factset, which isn't cheap but does seem to give good service
 
User avatar
gevor
Topic Author
Posts: 0
Joined: July 20th, 2006, 9:10 pm

Analyst data

July 10th, 2007, 3:40 pm

Thanks for the reply!Those guys?http://www.factset.com/And why/in what it is better comparing to IBES and Reuters data?
 
User avatar
csa
Posts: 0
Joined: February 21st, 2003, 3:16 am

Analyst data

July 11th, 2007, 5:15 pm

I've used Factset before, but it's more known for other stuff. First Calls/IBES is probably the most popular source, and whoever would test your strategy would also probably use the same data source - they don't have that much choice in terms of analyst estimates.
 
User avatar
gevor
Topic Author
Posts: 0
Joined: July 20th, 2006, 9:10 pm

Analyst data

July 11th, 2007, 5:20 pm

The thing is that I will test my strategy, and make the decision..So I wanted to do as good backtests as I can.So - thanks for the replies! If I found smth interesting, will write here.