Skip to main content
site map
my account
contact
our facebook page
Latest Posts

+
...

Market Timing

Sunday, June 09 2019
We have had questions about the recent MIPS two trades that went to CASH on 6/3/19 and back LONG 4 days later.
- Read on...
 
All mathematical models fail under certain circumstances.  For example, Auto Pilot software can fail in flight in some very "stormy" weather conditions, and the software must call for the pilot to manually take over.   And recently, we have heard about Boeing's software fail in one of the 737 MAX alerts (that they have not yet fixed).

In timing software models, an obvious problem is when a model uses daily data (say for the SPY ETF), and the market keeps changing direction every 3-4 days. Without intervention, this could lead to getting whipsawed because it can take at least 2-3 days for the model to identify a market change (like a new trend).. 

And, there are times when the model is not working properly, but the developers do not know why (so they can't fix it). For situations like this, a good model will have built in STOPS.  In general, STOPS are activated when the model is not working properly, but it does not know why.  There are no hard rules for when STOPS "trigger", nor when the investor should come back in.  For example, a common STOP is to go to cash when the SPY has dropped more than 5% from its most recent top (and the investor has lost 5%).  Problem is, it is easy to get out, but no popular rules exist on when to come back in. Using STOPS without effective rules can lead to large losses.
This is why MIPS has 6-8 built in STOPS (depends on which model) that have cut the "average loss on losing trades" by over 60% (see graph below or attached)
image.png


Back to Ground Zero...
We have had questions about MIPS' recent signals that went to CASH on 6/3/19, and went back LONG 4 days later....

The signal to go to cash on 6/3/19 came from a STOP in the MIPS model.. That particular STOP requires that the model does NOT issue a new signal for at least 4 days following the STOP signal. Normally, this would prevent large losses or at least go unnoticed, but in this case the market had 4 large back-to-back daily gains since the STOP (somewhat rare).  Even though that hurts, it can and does happen. But, normally it makes very little difference in the model's performance in the long-term (see 3+ year graph attached to put this in perspective). 

SEE ATTACHMENTS BELOW


Paul Distefano, PhD
CEO / Founder
MIPS Timing Systems, Inc.
Houston, TX
281-251-MIPS(6477)

support@mipstimiing.com
tond Zero...
have ha questions about MIPS' recent signals that went to CASH on 6/3/19, and went back LONG 4 days later....

The signal to go to cash on 6/3/19 came from a STOP in the MIPS model.. That particular STOP requires that the model NOT issue a new sat least 4 days following the STOP. Normally, this would prevent large losses or at least go unnoticed, but in this case the market hadback dewhat rare).  hurts, it does happen. But, normally it makes very little difference in the model's performance in the long-term (see 3+ year graph below (or attached) to put this in perspective).
Posted by: Dr. G. Paul Distefano AT 03:21 pm   |  Permalink   |  Email
 

MIPS Timing Systems
P.O. Box 691047
Houston, TX  77269

An affordable and efficient stock market timing tool. Contact MIPS
281-251-MIPS (6477)
E-mail: support@mipstiming.com