I've seen a number of "you get what you pay for" comments in regard to the Reuters and other subscription data feeds, and don't doubt that they put a lot of work into accuracy. But I ran a little test to compare the Yahoo EOD data I download with MLDownloader against the 5-year data CD that came with MS (through 5/31/05) just for fun. I used a stock I have a bit of familiarity with - ADPT. I've sucked Yahoo dry for historical data on all US stocks and as many indices as I can find symbols for, going back up to over 40 years for some stocks, so I have a lot of data on my PC. I pulled up ADPT's entire history from going public in the summer of 1986, then loaded the 5 years of ADPT data from the MetaStock CD and overlaid the two right on top of each other in different colors. Surprise! There were indeed discrepancies in the charts which are easy to see with different colors. Back in 2000 (the earliest data supplied with MS on the CD), the Yahoo data "blows up" and reflects old highs in the $40s rather than the $60s I remember and which the MS data shows. Hmmm, warning lights flash. But then I noticed something; after spring 2001 the data lines up again, exactly on top of one another. So I dusted off my brain cells and tried to figure out what was going on. Both datasets accounted for the three 2:1 splits over the years, but there was still that discontinuity through 2001. :eek: Then I finally figured it out; the exact date the data diverged was May 14, 2001 and guess what - that's the day Roxio/Napster was spun off and ADPT shareholders got shares of ROXI. The price divergence on that day between Yahoo/MS was about $13.50 - the value of the ROXI shares at that time. So, the Yahoo data had treated the ROXI shares as a dividend and subtracted their value from ADPT - which I think is the correct way to do it. The MetaStock data kept the pre-spinoff prices for ADPT. My conclusion is that, in this one case at least, the Yahoo data is MORE accurate than the Reuters data, and it's the free one. I wonder if others agree with this conclusion? I spot-check my Yahoo closing data every day after the download is complete, picking random stock symbols, just to check today's data against the last week or so, looking for obvious errors and discontinuities. So far, I haven't found any. Even if there is a random wild-day of data included due to database errors, it should be pretty easy to spot - it would be hard-to-spot systematic offsets and errors that would really get you into trouble by leading you down a bogus historical path that wouldn't leap out at you. Comments welcome from others about their successes and/or failures using Yahoo EOD data, and would especially like to find a source for all the index and sector symbols that are in the MS universe but don't work on Yahoo.

hey jim..... have noticed that phrase used a time or two myself.... even though it's not something i'd ever say, it is a concept that often rings true.... surely yahoo must take pride in the accuracy of their data, if for no other reason than to fend off legal hassles.... the download feature is a service that has been used for years by the excel crowd.... can't ever remember anyone talking it down..... to list all symbols do a symbol search and only put in a [color=red:1e6adb3c94] _ [/color] .... example link..... yahoo would never serve my needs today but it's nice to hear it suits yours.....h

If you look at the bottom of a page on Yahoo's site, you'll see where they get their data. It isn't Yahoo's data, they get it from CSI. "Historical chart data and daily updates provided by Commodity Systems, Inc. (CSI)." If I was going to go with someone besides Reuters DataLink, it would be CSI. They have great accuracy. A while back, I did a comparison of different data providers. I did speed and accuaracy. I even did accuaracy with different types of securities (Futures, Stocks, Indices, etc). I can't remember the results anymore, but Reuters DataLink was in the 1st or 2nd position in all the areas I tested. CSI was right there with it.

Errors in YAHOO data? Check the VOLUME for the Dow and the S&P 500: DOW: http://finance.yahoo.com/q/hp?s=%5EDJI SP500: http://finance.yahoo.com/q/hp?s=%5EGSPC --Johnathan

Hayseed, thanks for checking two other sources. I remember the spinout of ROXI from ADPT in May 2001; I was an ADPT shareholder. Maybe there's a question of interpretation; whether or not to adjust prices pre-split for the spinout value. As I recall, ADPT press releases at the time considered it a tax-free dividend to existing shareholders, but it's obvious from looking at Yahoo's data that it treated it as a split/dividend, adjusting prices prior to that date, and post-spinout Yahoo's data lines up with the MS historical CD. Any CPAs out there care to comment? Maybe it can be interpreted both ways.