Skip to main content
site map
my account
contact
our facebook page
+
...

Market Timing

Tuesday, 31 January 2017

The market pundits are citing things like the Trump rally has run its course and that February is NOT normally a good month in the stock market.  As "proof" they point out that, in the last 40 years, in the "good months" between November-April (where they all go when they come back from "sell in May"), February has average gains of only 0.06% (yes, that's six 100ths of a percent).  Furthermore, in the February's following presidential election years, the market has dropped an average of -1.85%.

So, what are the last few days in the market telling us from a technical standpoint.

See the graph below:
1) the bold black trend line on the left is the "Long-Term Trend",
2) the light gray trend line extension is what I have called the Post Election Trend Line, but it really
    is just a natural continuation of the long-term trend line.
3) the orange circle on the top right is a "fake break" to the upside,
4) the blue lines point out where the SPY has dipped to its trend line, hit it, and bounced back up
    (in technical terms, the support level "held its ground" each time, which is bullish).
5) For our hard-core technical MIPS members, let us not ignore another bullish technical indicator. 
    Today (Tuesday) the SP500 "candlestick bar" formed a near-perfect "up hammer", by opening down
    (-7 on SP500), falling another -7 point segment during the day (-14 total), and then bouncing back
    up to close +12 points above its low-point (-2 for the day).  PS - The graph below is for the SPY,
    not the SP500; but of course, the similarity is there.


Of course, today's action is not why MIPS has stayed in a long position for so long (read previous blog for why that is).  It is mainly because a good trend following model will rarely go short when the market is above its trend.  But, good trend following models (like MIPS) will have algorithms that will take themselves short or to cash with a relatively big drop above the trend line [e.g., reversion-to-the-mean algorithms in overbought (or oversold) markets].

Overall, by following the trend, the MIPS models have done well in the recent market (see graph below).
 - from TimerTrac.com


In the above, you can see the following results for the time period of Jan'16-Jan'17:

Red Line      - S&P 500 = +11.9%
Blue Line     - MIPS3 trading SPY 1.0x Long and 1.0x short  =  +11.1%
Gold Line     - MIPS4 trading SPY 1.0x Long and 1.0x short  =  +12.2%
Green Line  - MIPS4 trading SPY 15.x long and 1.0x short   =  +17.9% *
                     * note: 1.5 leverage on long positions


Stay tuned...
It appears that the market will continue to the upside, but let's wait for MIPS to provide us with more "good guidance" going forward !!!


Paul Distefano, PhD
President/CEO
MIPS Timing Systems, LLC
Houston
251-281-MIPS(6477)

Posted by: Dr. G. Paul Distefano AT 10:27 pm   |  Permalink   |  Email
Sunday, 29 January 2017

Trading frequency is a very important part of the performance of timing models.  If you look on TimerTrac.com, you will find long/short timing models that trade between 10-80 times/year.  If you use a long/short model, these "trades" are basically "round trip trades", which require two actual trades on every "signal change".  For example, a new long signal would require that you: (1) cover your existing short position, and then (2) buy your new long position.  At $10/trade, a model that trades 50 times a year (50 round trips) requires 100 actual trades (lots of work) and costs $1,000/year in trading fees. Whereas, a model that trades 12 times/year requires 24 trades/year and would cost $240/year at $10/trade.

As most of you know, the MIPS2 model trades about 8 times/year and the MIPS3 and MIPS4 models trade between 12-14 times/year.  Since MIPS3 and MIPS4 are better models, we recommend that you use them even though they trade a little more often.  Please remember that the trades/year are "averages" and could mean that in some years the MIPS3/4 models could trade 4-6 times/year and in other years they may trade 16-18 times/year.

What determines how many times/year a model trades is 90-100% dependent upon how long the market trends in that year last (other trades may come about from reversion-to-the-mean in overbought/oversold markets, etc.).  The MIPS models were developed to follow "Intermediate-Term" trends (weeks/months); and not "Short-Term" trends (days) nor "Long-Term" trends (months/years).

In the last 5-6 years, the market has trended up somewhat consistently.  However, in the graph below, you will see that the market in 4Q'15 trended down.  But, before and after 4Q'15, the SPY trended up in a relatively tight pattern.  So, after 4Q'15, models that were "long" more often than "short" should have matched or beat SPY (especially with like 1.5x leverage on long signals).




Getting closer to home, we all know that MIPS4 has been Long since 07/12/2016 (135 days)!
- [MIPS3 did basically the same, but with a one-day cash position on 11/04/2016]. 
Incidentally, in the 14 years since 2003, MIPS3 has had 11 signals that lasted longer than 3 months (90 days), with the longest signal lasting one year (361 days).  All 11 of these signals were winning signals (even a short signal in 2008).  The average gain from these 11 signals was 9.90% each.  [This would have been higher with 1.5x leverage, see the last graph.]

See the graph below to follow the trend of the SPY since 07/12/2016.  As you can see, the SPY trend was up most of the time, except for a dip in early November.  BTW, in a strong up market like this, it takes a bigger dip than that for MIPS to change direction.  And, chasing ralatively small dips can lead to getting whipsawed. Remember, our Blaster Series algos in 1Q'16 greatly lessen the chance of the MIPS models from getting whipsawed !!!


So, how did this 135-day long signal since 07/12/2016 perform. The graph below shows MIPS4 with a 7.9% gain and MIPS3 with a 5.8% gain, both trading SPY with no leverage.


 

With 1.5x Leverage
The graph below from TimerTrac.com shows how MIPS3/4 would have performed trading SPY with 1.5x leverage on long signals and 1.0x leverage on short signals (MIPS4 +11.8%;  MIPS3 +8.6%).


 

Paul Distefano, PhD
MIPS Timing Systems, LLC
Houston, TX
281-251-MIPS(6477)
www.mipstiming.com


Posted by: Dr. G. Paul Distefano AT 11:40 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