Jun

19

As we all know, the market has had two small up days but is down over longer periods starting with 3 days. Yesterday's action was unusual in that it was a big up opening, but down consistently from the opening. I find that such a condition has never occurred before, although it all seems quite reasonable and de rigeur. What other seemingly normal things hardly ever happen, thereby violating the random walk?

I dare say an artful Downian simulation of the last 3 days would show that the actual transpiration of events was quite normal under random numbers, ( the magnitudes of the opens would have to be taken into account), yet it didn't in real life. Does it signify anything except the infinite creativity of the market mistress, who is so likely to find that those louses will go back to their spouses in these terrible days.


Comments

Name

Email

Website

Speak your mind

5 Comments so far

  1. Tom on June 20, 2011 12:16 pm

    Dr Niederhoffer,

    Please consider May 19, 2004 as being an example of “a big up opening, but down consistently from the opening”, in a similar location in the overall market cycle to June 17, 2011.

    This marks a change in cycle, the same as I believe today has confirmed. Note in both cases the new bull market, the change in public sentiment to go long, followed by a bearish correction lasting around six weeks, with public sentiment now being rather nervous and plenty of uninformed short selling activity.

    Filtering merely by the criteria of a big up opening, but selling off from there and closing near the low, we can add 9/8/8, 9/19/8, and 10/14/8. However I recognise that both the general market condition at the time and the function of these movements is different.

    I think the more meaningful comparison is between 5/19/4 and 6/17/11.

  2. vic on June 20, 2011 4:26 pm

    one has considered what you say and do in part believe it. but one sparrow dont make a spring. vic

  3. Tom on June 22, 2011 2:32 am

    We can see by how price action has been unfolding these last three sessions some evidence of a cycle change which supports the earlier reasoning.

    I agree 100% with your comment about sample size. To me, this is one of the biggest challenges in trading, and goes directly to methodology,

    Do we construct a hypothesis based on what we reason about the market (what seems logical to us based on what we understand or think we understand) - and then look for the market to prove or disprove by looking at market behaviour (ie price).

    Or do we look for correlations and patterns in the market, which gives us more evidence in a scientific / statistical sense, but suffers issues both with sample size and from flaws in the approach. Namely lack of context and explanation for factors that drive price - lead us to lump together a bunch of conditions which look similar but are not similar.

    For example, using just pattern recognition alone, we could lump all the Sep/Oct 2008 strong opening + selloffs together with the 2004 and 2011 ones, when it would be incorrect to do so based on context.

    There must be some great patterns / indications / edges / etc in the market which can be exploited. In some cases they will be rare enough that we can’t expect the to reoccur frequently enough in our lifetimes in one asset class to give any statistically significant results. So how do we overcome the “one sparrow” problem?

  4. Tom on June 23, 2011 4:08 pm

    Price action today (weak opening, closing almost unchanged, 20 points off the low) is further support of the cycle change contention from last week.

  5. Tom on July 2, 2011 7:09 am

    Is the action of the market over the last two weeks enough to offer proof of the cycle change?

    Bacon wrote that the public misses the changes in cycles. It would appear to be very profitable indeed to avoid doing this. Aside from this isolated example, is it possible to consistently predict changes in cycles?

Archives

Resources & Links

Search