Barrons' 2015 Picks on a Weekly Basis. 12.17.15

Last week we looked at how the M11 model improved Barron's top 10 stocks' performance.
Here's the same portfolio looking back 7 years and traded on weekly bars instead of the M11's usual daily bars.  Right away we notice that the daily bars do provide a superior return when looking at the net 1 year returns for the 2 models. (both the daily and weekly models are set to mean reversion mode)  The weekly model also uses a top 2 sort in lieu of the daily version's top 4 sort.  Then we have to consider the commission costs for daily versus weekly models and the amount of time required to maintain a daily versus weekly rotation..
Nevertheless, there's a couple lessons to be learned here including finding a suitably attractive portfolio to apply the M11 filter and then rotating paradigms accordingly as the market dynamics favor either a momentum or mean reversion based tactical approach.

M2 Trader Progresses....12.16.15

It's 60 minutes past F Day and the FED has finally raised rates for the first time in 7 years.
The initial reaction was bullish but we'll check back on Friday's close to see if the enthusiasm held. These are the 2 most bullish weeks of the year historically so there's that inertia going forward.
I've made some major changes and refinements to the M2 project over the past few days, in the process re-learning EXCEL programming that I had long forgotten.
The major effort has been focused on automatically calculating the limit stops, which is really the most important part of the program as it calculates and applies current volatility (risk) thresholds to each position. There's a lot of volatility analysis now embedded in the model and the stops are calculated based on 2 separate set of metrics.  Just to make matters more confusing a programmable longer term risk tolerance adjustment factor is included so various "what if " risk scenarios can be
quickly examined.  The M2 dashboard also now indicates whether the current mode is momentum (MN) or mean reversion (MR). Needless to say, the M2 project has been a long time coming and involved much more time and effort than I imagined when I started. M2 is just about ready for prime time although I still need to run some quality assurance tests before releasing.
Thanks to all who offered input ideas and suggestions on making the program what it is.
This is the SSO/SDS version (momentum mode) Top 1...NOT delta neutral which is Top 2..
Probably have to click once on dashboard to enlarge abd see details.

Ponzo Updates for SPY and the VIX Index....12.15.15

It's almost as if the Ponzo charts are giving us a contrarian view of the markets based on last week's bullish outlook.  And then this week we're looking considerably more subdued in the forecast.
Keep in mind these forecasts are based on 25 year lookbacks and a lot's happened in the markets over that time.  Those who watch the markets in real time on a daily basis know how the HFTs totally controlled the ebb and flow of price.  This is particularly evident on high volume days like Monday where we saw quarter percent reversals on 2 minute bars, which were then reversed within the next 3 bars.  Yes. the VIX is in the 20s but that kind of action in the big ETFs like SPY. QQQ and IWM is all HFT driven and the trick is to avoid the collateral damage of the whipsaws.  
It's interesting that the standard deviation of the VIX is about 4 based on a price of 20, whereas the SD of SPY is 6 based on a price of 205. Let's hope we don't realize that VIX yellow line forecast (or the SPY yellow line).

A Weekly Version of M11...12.14.15

This is a version of M11 using weekly bars so you only have to rotate positions once a week.
As with the old T2 model if a position is stopped out you just wait until the following Tuesday AM to open new positions. We have to set the stops a bit higher than the daily model since we are covering risk for a week (on a cumulative basis).  The M11 portfolio is the default daily model and the limit stops are arbitrarily set at 2% for all inputs.  This is the mean reversion mode.
Note the highlighted 7 year returns and the differential risk exposure of SPY against M11.
This is a top 4 sort which works fairly well in conjunction with the full beta spectrum inputs.
Click once on dashboard to enlarge.>>>>>

VDX Updates for SPY, XLU VIX and TLT....12.12.15

Looking forward to the historically bullish next week we're coming off the worst week since August and certainty not tracking the latest Ponzo forecast, which suddenly turned bullish ;last Monday. Currently SPY is below the 200 and 50 day MAs and extremely oversold by any technical standards.
The VIX is now wildly overbought after a 30% rise on Friday. In the past one day surges in the VIX that exceeded 20% were destined to produce a positive day for SPY 90% of the time.
We'll just have to wait to find out if Monday's close yields the expected results.
We are suspending coverage of FXI (China 25).  The highly manipulated Chinese markets and the questionable validity of state issued economic and policy data make China unattractive for now.
Next week, in case you've been in an isolation booth, the FED is expected (86% Fed Fund futures consensus) to raise rates by .25% and opinions are varied on whether this long awaiting change in FED policy will create more or less volatility in the markets.
Work continues on the XX volatility model including the limit stop indexing routine.
See Fridays; M3 Insight.

M1 Volatility Model Progress....12.10.15

We're chugging along with improvements to the volatility model including 3 algorithms (ATR, SKEW and Median) that track both 30 day volatility using a spectrum of metrics.  I generally favor a visual data display and have used pattern recognition for years to gauge market odds.
Looking at the 3 lower new volatility based charts within the M1 dashboard we can quickly see that both the ATR and the PCL (previous close to today's low) are increasing.
The third chart on the lower right has a more benign pattern, displaying a mere .76% delta for the median day open to close range. The obvious conclusion is that the big action occurs overnight and this helps us to craft the limit stops accordingly since we now know the likely intraday range swing.  
M1 is still a work in progress but these latest metric additions will facilitate gauging the stop odds.

Mosaic filters Barron's top 10 Picks...12.09.15

Many savvy investor &.traders read Barron's magazine each week.  They're running a promotion right now that's pretty attractive. Some articles are free, some are locked if not a subscriber.
This week they selected their top 10 stock picks for 2016....and also looked back to how their top 10 picks for 2015 have performed., which turns out not so well, with a net loss of 6% for the year in an equal weighted portfolio of the 10 stocks.
Here then is how that same portfolio would have fared using M11, a top 2 sort and our default stops.
Yes, there would have been some slippage along the way because some of these stocks had really bad days that would have blown through our stops and commissions are not included.
Nevertheless, M11 is designed as a starting point, not a final solution and even after handicapping our model by 20% for the vagaries of the market an 80% gain looks a lot better than -6%.
I've also loaded up Barron;s picks for 2016 on the M11 platform and set a top 2 sort.
We'll check back each quarter so see how M11 is performing relative to Barron's passive strategy. Past performance is no guarantee of future results but this looks a case where a $ 50,000 portfolio of Barron's top rated stocks would have produced somewhere in the neighborhood of a $40,000 return using M11 as opposed to a loss of 6%,
Hint for accumulation:  buy on a 3 day low for the SPY or when at lower oversold levels per VDX.
This is why I believe in technical analysis and active money management.

And looking forward >> Here's how Barron's picks for 2016 would have performed in the past.