| Index | Recent Threads | Who's Online | User List | Search |
|
|
 |
Portfolio123 » List all forums » Forum: Errors » Thread: Analyst Rev Field for Ticker=SWIR |
|
Total posts in this thread: 5 |
[Request a Feature]
[Post new Thread] [Add To My Favorites] [Watch this Thread] |
| Author |
|
|
grokkalot
Advanced Member
|
In one field of a ranking system, I'm getting a low value for the formula IsNA(NextFYEPSMean-NextFYEst4WkAgo,0)/Close(0) applied to the ticker SWIR. The value implies that NextFYEPSMean must be less than NextFYEst4WkAgo for SWIR, but neither Reuters website nor Yahoo agrees with that statistic. Can you tell me whether this is indeed the case in the fundamental database on June 9, 2008 and, if so, why the difference from what is shown on Reuters website: http://www.reuters.com/finance/stocks/estimates?symbol=SWIR.OQ |
||
|
|
marco
|
They are both 1.62 which matches Reuters values for the row "Year Ending 09" To quickly check values simply enter the following rules in a new screen Rule1: NextFYEPSMean Rule2: NextFYEst4WkAgo Rule3: ticker("swir") Then select "Screen Factors" in the Report drop-down and click "Run" ---------------------------------------- Portfolio123 Staff |
||
|
|
grokkalot
Advanced Member
|
Thanks. Using that technique, I was able to confirm that the formula I gave above also evaluates to zero, as expected, so it is not a data error. Yet the reason I was investigating this is because I didn't expect that a formula value of zero for this fundamental attribute would result in a ranking from the overall universe (including lots of stocks with no analyst coverage at all) of 18.89. On the face of it, that means that 81% of stocks in the universe of all stocks had analysts revise up their next years earnings estimates in the last 4 weeks. Is that right, or is this an artifact of lots of stocks evaluating to zero for this formula. Let's say that for a given formula, 10% of stocks evaluate to a positive value, 10% evaluate to a negative value and 80% evaluate to zero. What do we expect as the rank value for stocks with a formula value of zero for that factor? Is it 50 or 11? |
||
|
|
marco
|
The actual values are NextFYEPSMean = 1.619 NextFYEst4WkAgo = 1.6200000000000001 Looks like rounding is making this stock slightly less than 0. Since the factors you chose have a high number of NA's (probably over 50%) it evaluates to 0 most of the time (due to your IsNA). All the 0 values are bunched together and get the same rank. A number slightly less than zero gets the next percentile value and 18 sounds about right. I think a simple mathematical Round function is needed. ---------------------------------------- Portfolio123 Staff ---------------------------------------- [Edit 1 times, last edit by marco at Jun 9, 2008 11:13:13 AM] |
||
|
|
grokkalot
Advanced Member
|
Strange...I tried the screen you suggested this morning with the equation IsNA(NextFYEPSMean-NextFYEst4WkAgo,0)/Close(0)=0 and ticker("swir") and got 1 item returned, and now it has changed so that it only returns the 1 item with the inequality IsNA(NextFYEPSMean-NextFYEst4WkAgo,0)/Close(0)<0. My question above would have been different if I hadn't been successful retrieving swir from the screen using the equality form. I guess what I would really like to do here is have the ability to do a sum of z-scores rather than ranks in the immediate parent node. Anyway, thanks for your help, and I am still interested in the answer to the question I asked immediately above (is the rank in my artificial example 50 or something close to 10?). |
||
|
| [Show Thread Printable Version] [Post new Thread] |
