Bear Mood Increases and more MVP Studies.....05.11.16

Half hour till the close and so far we're looking at an ugly reversal of yesterday's surge with discretionaries (XLY) taking the biggest sector hit.  I came across this Bloomberg piece on the rise of the bears  which goes hand in hand with previous links to the massive increase in VIX calls in anticipation of the Sell in May herd mentality that will likely drive the markets lower for a bit....perhaps through June, which is historically a weak month anyway..
The VIX is up over 6% today and the pause below 14 was short lived, to say the least.
Continuing the MVP studies here are the MVP trades only using the ATR and PCL versions.
For ease of study the 2 year unstopped benchmark price and the MVP trade results are shown on the same chart.
At least 2 observations should be clear.
1.The ATR returns could be substantially improved by use of a trailing stop (see yellow arrows on the equity chart).  Giving back 2-4 % after a nice gain is not necessary with simple stop controls.
2. The PCL returns are much better than the ATR returns and it all has to do with the relative value of the PCL algorithm versus the ATR.  Note that we do not have the trailing stop issue with the PCL trades in the same manner at the ATR.
The green dotted vertical lines designate the actual trades in the ATR and PCL models.  They are not the same....and again demonstrate the relative attractiveness of the PCL approach.
While the MVP remains a work in progress we can now see how only a few very high probability trades can produce decent returns while keeping capital securely in cash the vast majority of time.

 

PONZO Updates + More...05.10.16.

This week's PONZO updates have reversed the previous bullish outlook for SPY and set a more defensive to neutral forecast at least short term.  That's pretty much the case with TLT also, continuing the divergent decoupling that we normally expect from this pair.. The VIX outlook remains benign and with today's close we're back below 14..a situation we forecast over the weekend. Just for amusement here's a piece from Bloomberg on more Chinese market insanity. 
Will traders never learn from the monumental mistakes of the past?

MVP Prelim Studies....05.09.16

My initial MVP study models using the ATR and PCL as buy triggers are shown below using the M1 platform .  Just as a starting point for further research I've simply replaced the short term momentum and mean reversion filters with either an ATR or PCL threshold reversal.  The reaction is similar to that of a mean reversion filter....that is.. the signal looks for a 2 day pullback form a high threshold and then triggers.  I've made the threshold adjustable according to near term volatility so we get more signals than just the 11 that would be expected based solely on threshold level violations shown last week.
This is just one example of how we can use the ATR and PCL effectively to mitigate risk and raise returns above a buy and hold strategy. Later this week I'll profile a couple variations of this theme.
The best returns are likely to be achieved by trading short once a ATR or PCL threshold trigger is violated to the upside, but that's difficult to test using the M1 platform....for now..  
The PCL version performed somewhat better than the ATR in this study but in both cases the unstopped raw price pattern of SPY is much improved using either filter and the drawdown dips are significantly minimized as detailed on the 2 year MVP Equity chart (ATR study shown)...

   

VDX Updates & Weekend Reading...05.08.16

This week's VDX updates suggest SPY in an oversold condition and TLT approaching overbought. Finding a trend has been difficult lately as earnings reports have created the usual surge of volatility associated with earnings uncertainty.  The VIX has been particularly squirrely, as noted in the Bloomberg article link above, and, discounting significant negative news, is likely to retrace back to the 14s within the next 2 weeks as the Sell in May VIX premium pop mentioned earlier in this week's Barron's post is adjusted back to a neutral skew.
Going forward we'll continue to explore less traditional technical signal parameters such as the PCL asnd ATR in an effort to constraint risk exposure in the face of deteriorating global economics.
Monday we'll present the latest performance metric studies of the ATR/PCL indicators, which now include a limit stop function.  
There's still considerable work to compete before these platforms are ready for prime time but the prelim results are more than encouraging.

MVP Charts, a Mixed Bag....05.05.16

Today we continue the MVP studies, including XIV, the VIX inverse ETF, that's been difficult to trade this year for those that have relied on a strong XIV/SPY correlation.
As mentioned yesterday the MVP is a different way of looking at integrated price and volatility dynamics and although the frequency of trades may only be a half dozen a year, the reliability of the signals is very attractive and the rest of the time capital is in cash..
Note that the different ETFs have different trigger dates and frequency of trades.
Best to click on each chart to see detail better.
If you're trying to mimic the ATR and PCL charts by tweaking the M1 platform be advised that the "Trend" algorithms are not just simple moving averages, but more elaborate constructs, so your signal intersects may not match those shown below..
July 4th weekend the MVP platform will be available to Mosaic subscribers.

And, finally XIV >>> Note that the trigger thresholds are up at 6%+, reflecting the high beta of XIV relative to SPY and the other index ETFs. The ATR/PCL offset is also greater than the other low beta ETF studies to compensate for the resultant higher lag between the 2 indicators.

MVP Forecasting Refinements....05.04.16

As the VIX heads towards 17 today and into overbought territory our work continues on the MVP (Mosaic Volatility/Price charts.  Some of the recent refinements are shown below in a study of DIA (DOW30 proxy ETF).  Although it theoretically reflects DOW price dynamics the DIA is actually an anemic little ETF with only marginal daily volume and more surprisingly perhaps not one member firm of HedgeCo,  (a consortium of some 1400 hedge funds) owns a single share of DIA..  They own plenty of SPY, IWM and QQQ but regarding DIA...nada.!  Why DIA is shunned by most funds is a story in itself but suffice to say that since it includes only 30 stocks it is the target of massive intraday manipulation by large prop shops...both domestic and foreign and therefore generally hated by risk managers, Nevertheless, if we play the long game using MVP it is clear that there are occasional high odds vesting situations when the ATR and PCL volatility peaks achieve signal threshold levels.  The MVP approach is completely different from the standard M1/M3 ranked rotation methodology and the MVP signals can be few and far between, depending on the ETF or stock under study.  When generated, however, the signals are consistently reliable, which is what makes them attractive.
Note that the green dashed vertical tracking line is NOT vertical...it is offset by 4 days from the ATR to the PCL to accommodate the leading/lagging behavior of the 2 signals..
Tomorrow we analyze various volatility/price correlations in more detail.

Ponzo Updates as Global Growth Fears Shiver the Markets...05.03.16

Another day, another worrisome flurry of economic reports that do not bode well for the markets.
After a first day of the month rally that looked like it might have some legs sour IMF reports dropped the markets overnight and the selling continued into US market hours.
Bonds were the big winner as flight to safety became job one for risk managers. TLT is the best performing of our M11 bond basket followed by TIP....both ETFs have good volume but serious bond traders should keep an eye on JNK and HYG, which look ready for a possible breakout and have volume interest 2 to 4 times that of TLT..
The Ponzo updates this week closely mirror last week's updates so no new trend forecast changes at this time.
After digesting 6 separate well argued articles on the Sell in May thesis the best conclusion I can come up as a risk manager is that May to October is not necessarily bearish...it's just not bullish.