yeh, read it and it doesn't say that. It just like doesn't. It says the opposite, and sparked a mini-industry in writing papers attempting to prove the proposition that OJ futures are a good forecasting device. There were so many of these damn papers that the urban myth grew up that the original Roll paper was one of them. Roll's actually quite a good guy, very thoughtful for a finance professor. It was also him that came up with the "Roll critique" that you mentioned earlier - that market efficiency is an unverifiable concept because any empirical test of it is a joint test of EMH and the assumed data-generating process.
I did a bit on this for my website a while ago and the ref I dug up was http://papers.ssrn.com/sol3/delivery.cfm/SSRN_ID396645_code030519500.pdf?abs tractid=396645 . The piece is at http://www.crookedtimber.org/archives/000340.html btw if anyone cares. dd -----Original Message----- From: PEN-L list [mailto:[EMAIL PROTECTED] Behalf Of Doug Henwood Sent: 24 June 2004 03:20 To: [EMAIL PROTECTED] Subject: Re: Marxist Fianancial Advice Daniel Davies wrote: >That point (which is, incredibly, very well established and true) is that >40% of the entire volatility of the NYMEX contract in September Frozen >Concentrated Orange Juice occurs in the single day on which the Department >of Agriculture releases its forecasts for orange production. I just can't >for the life of me see how this factoid is at all consistent with the FCOJ >futures market having any information advantage over the US government. Surely you've read, or read about, the paper (I think by Richard Roll) claiming the OJ futures market is better at predicting the weather than the U.S. Weather Service? Doug
