| Index | Recent Threads | Who's Online | User List | Search |
|
|
 |
Portfolio123 » List all forums » Forum: Data related issues » Thread: Bug in BV5YCGr% for GIS |
|
Total posts in this thread: 6 |
[Request a Feature]
[Post new Thread] [Add To My Favorites] [Watch this Thread] |
| Author |
|
|
jtbaccarat
Advanced Member
|
There may be a bug in the calculation of BV5YCGr% in this example for GIS, attached. ---------------------------------------- 
P123 GIS.PNG
(10888 bytes)
(Download Count: 59)
|
||
|
|
marco
|
Looks like Reuters has had this value for GIS for quite some time. What's your source? ---------------------------------------- Portfolio123 Staff |
||
|
|
jtbaccarat
Advanced Member
|
Hi Marco, My source is simply taking those values and calculating the IRR, which is close to 10%, not 145%. I am not sure of Reuters' internal calculation, but I use IRR using the final and initial values, which is close enough to the same factor for other stocks. 145% is way off. I included the values needed for the calculation in the PNG. Use a calculator, Excel, or a formula like this: ((Final/ Initial)^(1/T))-1 Thanks for taking a look at this. |
||
|
|
marco
|
Looks like they are using annual values: On the last 10K for 2006 the BV was 5,772 On the 10K for 2001 the BV was 52.20 That's 156% which is close 145%. I'll find out from Reuters about the difference. The company is now on Q2 of 2007 with a BV of 5,546. At the end of Q2 of 2002 the BV was 3,552.70 If Quarterly data was used the growth rate comes out to ~9% Once we start calculating our own ratios we'll most likely calculate the 5Y BV growth rate using Quarterly data. ---------------------------------------- Portfolio123 Staff |
||
|
|
marco
|
I think calculating the 5Y growth rate of the BV using annual reports is probably the correct way. Annual reports are scrutined and audited more carefully than Q's, which is probably why Reuters decided to calculate it using Annuals. In your case the confusion is from the fact that GIS's BV jumped from 52 -> around 3,500 soon after 2001's Q4. If we had BVPSPY5 we would probably see a very small number. ---------------------------------------- Portfolio123 Staff |
||
|
|
jtbaccarat
Advanced Member
|
Hi Marco, Thanks for looking into it. It will be great when we are able to calculate the historical ratios; that will be an amazing improvement; and, also recalculating ratios such as this example (there's always more than one way to calculate these ratios). On another note, I was trying previously to recalculate a few factors but I found the formula box was too small. One specific example: Surprise%Q1, etc. seem to have -ves divided by -ves giving +ves, so it seems to give misleading results; therefore, I tried calculating Surprise in a formula, but quickly ran out of space. Any chance these formula boxes could be increased a lot more, or is there a reason that precludes the increase in this boundary condition? |
||
|
| [Show Thread Printable Version] [Post new Thread] |
