[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: FW: Random walk



PureBytes Links

Trading Reference Links

OK.  I'll bite.  as a psychologist by training I learned than all data set have 
variation.  duh.  The trick in developing knowledge of some system (human 
behavior, markets, manufacturing process) is to be able to account for some of 
that variation by controlling and then varying some idendpendent variable.  My 
parametric statistics model establishes some confidence level that the 
variation I have accounted for is not random (for some reason .01 and .05 
levels of confidence have been adopted.)  

Yale Hirsch says Monday's are on average up days since 1980.  If that bias is 
sufficient statistically at say the .05 level of confidence, we can say the 
markets behavior on Mondays is 95% certain to be non-random.

So perhaps among other things, the closest we can get is a statistcal 
probability that events are random/non-random.  Personally, I don't think 
anything is non-random.  Some things just have such complex causality that we 
can't (yet) explain them.

Quoting Brent <brente@xxxxxxxxxxxx>:

> Interesting Eric, have you tested any data with a random trade generator
> that proves that money management is more important?
> 
> I did a lot of testing at one time.
> 
> What details are you interested in.
> 
> I have also tested the idea that the 1st and last hour of trading mean
> something.
> 
> Really, the market isn't able to absorb sudden increases in volume. I guess
> I should ask what market, where and when.
> 
> I'm kind of disappointed, I thought someone here could come up with a good
> definition of non-randomness.
> 
> It's kind of fun though to see some persons get really bent out of shape
> about it though.
> 
> Brent
> 
> 
> 
> 
> 
> 
> -----Original Message-----
> Wrom: WOYIYZUNNYCGPKYLEJGDGVCJVTLBXFGGMEPYOQKED
> Sent: Friday, January 10, 2003 12:41 PM
> To: Brent
> Cc: OmegaList
> Subject: Re: Random walk
> 
> 
> A random trade generator illustrates that money management is more important
> to successful trading than good entries.
> 
> Care to share some details?
> 
> On the otherhand, I would suggest 1st and last hour of trading day are
> intent driven = non-random.
> 
> The market is not able to absorb the sudden increases in volume, and so
> there is pressure in a particular direction.
> 
> Institutional traders will act both at particular times and on some days
> more than others.
> 
> Eric Svendsen
> 
> 
> 


Best Regards,

Jim Johnson