February 12, 2012. The basic trend indication remains positive. Last week's correction has now balanced the internal structure of the market to the point that, if the underlying rising trend remains valid, it could well reassert itself this coming week, as the oscillators have now returned to neutral.

Accordingly it should be just a matter of days until our short-term study (KP-2) switches to "Plus 2" — maximum long exposure.

February 4, 2012. The KP Model rests on several conceptual building blocks. Two of these are: (a) that our inputs are driven primarily by internal market data, not prices, and (b) the Model gives additional weight to the behavior of the more speculative areas of the market.

After a tepid start In the first few weeks of the year, these more speculative sectors (small caps, biotech, emerging markets, etc.) have now come to life, driving the Model Line deep into positive territory (see chart). In the process, the markets have once again become overextended on a short-term basis and are, statistically, due for a correction. Thus, our mean-reverting short-term study (KP- 2) remains at "Plus 1."

January 22, 2012. The KP-1 Trend Model shifted to a "Long" status on the close on January 9 at 1281 on the S&P500. However, this recent shift toward more speculative asset categories has been accompanied with some restrained internal momentum as can be seen on our oscillator chart. Our short-term overlay is at "Plus 1" — bullish, but with some restraint.

January 8, 2012 . The KP-1 Trend Model remained in a negative status last week, but any further internal strength going forward will likely shift the Model to a "Long" status next week, with a short-term exposure of "Plus 1" (overbought = moderate positive exposure).

As noted last week, the US equity markets appear to be converging on a decision point -- a point where either the downtrend that began with the market's top last May might resume, or where the market indices continue to move upward beyond recent highs — an event that would signal the beginning of another significant "leg" upward, similar to the 180-point S&P move last October. In other words, we have arrived at a "risk-on/risk-off" intersection of some importance.

The lethargic action of the Model in recent weeks has been suggesting that the next major move would be down, while the price action itself (for example the relative weakness of consumer staples issues and the relative strength of financial issues) has been more constructive. Over the years, we have found that the internal market data that form the foundation of our Model have more predictive content and should be followed at all times. It should also be noted that this reluctance of the Model to align itself with the recent market strength is not a function of the recent Model adjustments; the previous version of the Trend Model has also remained in a defensive status through this period.

As always, we will provide clients with daily commentary as we approach this potentially important inflection point.

January 2, 2012 . There continues to be sufficient internal weakness to keep the KP Model in a negative or "short" trend status as of the end of last week, and the short-term condition of the US equity markets is now overought, prompting the KP-2 exposure control study to remain at "Minus 2," — maximum hedge.

The internals are neutral and converging as can be seen in our oscillator study, suggesting that the markets are near an important decision point.

December 26, 2011. The KP Trend Model remains negative and, in fact, demonstrated little response to last week's rally back to recent trading highs in the S&P 500 (around 1260). This divergence suggests persistent internal deterioration. Thus, while the period immediately ahead may have strong seasonal tendencies in its favor, the internal conditions monitored by our work suggest that we might be coming up on a negative surprise.

As the markets rallied back toward overbought, our exposure control study (KP-2) shifted to "Minus 2," — maximum defensive. In most applications, that suggests a net short position.

December 17, 2011. With last week's weakness in US equities, the KP Trend Model fell further into negative territory. It is noteworthy that the Model beame negative on November 23 and barely responded as the S&P 500 rallied 100 points (from 1160 to 1260), but made a new low as the S&P fell back toward 1200 last week. That kind of behavior is called a "technical divergence," and is usually resolved in favor of the market internals — in this case our Model Line.

By week's end, the markets had become sufficiently oversold to shift our exposure control (KP-2) to "Minus 1," which we implement in our Rydex models by shifting our short exposurre to cash.

All charts on the website now reflect the recalibrated models announced several weeks ago.

December 10, 2011. Our Model remains negative although the markets will need to decline this next week to remain in that position.

In recent months, we have completed a major project of recalibrating our Model in order to better manage the significant increase in the volatility of the equities markets, which we expect to be with us a for a while. These parameter adjustments have reset the number of annual model changes of direction to the traditional area of 5-7. A full description of these changes and their expected impact can be found in our December 1, 2011 Client Update.

November 27, 2011. The Trend Model is negative and our Exposure Overlay remains at Minus 1, suggesting limited exposure to the hedge (the Model prefers cash rather than short exposure).

November 23, 2011. On the close of Tuesday, November 22, the KP-1 Model shifted back to anegative trend outlook. The KP-2 overlay status, however, stood at "Minus 1," suggesting less than full commitment to hedge or short exposure given to the signifcant oversold condition that has accompanied the Model shift.

That oversold condition remains in place, especially given Wednesday's continued weakness. Statistically, a sharp short-term rally is quite likely.

The model adjustments we have discussed with clients over recent months will be implemented next week — by month-end, November.

November 19, 2011. The Trend Model remains positive and the recent corrective price adjustments in the indices has now moved the KP-2 exposure conrol to "Plus 2" — maximum long.

November 13, 2011. As discussed last week, the consolidation of the market indices around current levels caused the exposure-controlling KP-2 study to briefly dip to "Plus 2" -- maximum long exposure, however, the study remains closely balanced and shifted back to "Plus 1" on Friday. The primary trend Model remains positive.

November 6, 2011. The KP-1 Trend Model remains solidly positive, despite the recent corrective activity around 1250 on the S&P 500, +/- 50 points. The short-term work is mildly overbought, but it is quite likely that any further consolidation will cause our KP-2 signal to move to "Plus 2" sometime later this coming week.

We are close to completing the re-examination of the last 15 years for this market model, dividing the period into three 5-year slices: 1996-2001, 2002-06, and 2007-11. Obviously, the most recent five years are of greatest interest since we believe that the underlying character (what we call the "signature") of the securities markets has changed. However, examining all potential enhancements of the system across the entire database provides a safeguard against over-optimization (i.e., data fitting to the period under study).

The volatility-driven circuit breaker (a process by which deeply oversold markets cause the Model to switch to "cash") discussed last week has now been fully evaluated and found to add value across the spectrum of the last 15 years, although it has been invoked far more frequently recently. We are also examining lengthening some of our smoothing parameters to distance the model a bit more from the commodity-like "spikey" price behavior that seems to have become a recurring theme in the last 4-5 years. All of these modifications to the model will be implemented in coming weeks, and clients will receive an updated narrative guide.

We continue to believe that trend analysis is an indispensable prerequisite for the professional management of investment portfolios, as the prices of individual securities and industry sectors remain highly correlated with market indices, and of course, with each other. We do not necessarily predict more financial chaos but cannot imagine going forward for the next few years with a traditional long-only portfolio structure.

October 31, 2011. With last week's burst of strength in response to the Euro-finance package (Thursday), our Trend Model has now achieved considerable distance from its neutral (buy/sell) line, reflecting the development of significant momentum.

While the near-term is most likely to experience further correction of that move, we expect the Model to remain positive for the forseeable future.

October 22 , 2011. The return to a "Buy" status is now two weeks old and, while still young, appears to be developing some upside momentum as we can observe on the Model Components chart.

We have spent considerable time in the last few months examining the possibility that the internal structure of the markets may have changed significantly since the 2000—2004 base period when we calibrated this KP Trend model. For one thing, the frequency of changes in direction has increased steadily from 5.5 changes per year in the base period to 8.0 in the last four years, and 12 thus far in 2011.

The day-to-day volatility of popular market indices, particularly the volatility of smaller-capitalization indices, also appears to have doubled over the last few years. No doubt the expansion of computer-driven High Frequency Trading (HFT), has played a role in this apparent change of market personality. Although the time horizon of all these nano-second algorithmic strategies is less than one day, they do seem to cast shadows that can extend for several weeks.

We have examined all of this in considerable detail and have added two new groups of decision rules to the Model: (1) a "speed bump" which temporarily moves the trend signal to "cash" when the Model reverses itself too frequently; and (2) a volatility-driven circuit breaker that also moves the Model status to "cash" when relative volatility surges dramatically over a short period of time usually near the end of sharp declines. We expect these changes to add value to the process going forward.

Significantly, our analysis found the underlying logic of the KP-1 processing engine to be quite robust and not in need of significant adjustment. We could say that we have replaced the tires and added some new headlights to our car, but that the base vehicle remains unchanged

October 16 , 2011. On Monday October 10, the KP Trend Model did indeed return to a positive status and may now be developing some momentum behind the new bullish trend signal, after several weeks of frustrating reversals.

Our short-term work is very overbought and suggestive of at least a small correction, but we would point out that several other instances when the green KP-2 oscillator has reached these levels, the markets were in the initial stages of a major new advance. We are open to this possibility.

October 8 , 2011. The KP Trend Model remained in a "short" status last week, as prices generally advanced. Once again, the model has come very close to crossing its "early entry" trigger, as can be seen in this chart.

Since our model employs a variety of filters and smoothing techniques the Model components can develop a certain momentum of their own, and that is the case at the moment. Barring a significantly negative day on Monday (Columbus Day holiday, but the NYSE is open), the trend model will revert back to a "Long" status on the close. We will advise clients in mid-session.

October 2 , 2011. The KP Model's signals have been quite erratic in the past few weeks, with the decision lines very close to each other. At the end of last week, the Model's status was "short."

September 25, 2011. The KP Model, after an unproductive six-day shift to "Long," returned to a "Short" status on the market close last Friday (September 23). As we analyze all the data, the KP Model could remain closely balanced for several days, as was the case at the time of the previous week's shift.

Sharp, sudden price reversals do not treat trend-following systems kindly, but these reversals, or whipsaws as they are called, are the toll that systematic risk managers are required to pay in their pursuit of being on the right side of the major themes.

With the model closely balanced, we will communicate with clients as we consider it necessary, beyond our daily e-mail status reports.

September 18, 2011. On Thursday September 15, the KP Trend model shifted to a positive status at 1209 on the S&P 500. At the time of the change, our short-term studies were heavily overbought, bringing the KP-2 calculation to Plus 1 (moderately aggressive). These short-term oscillators remain at this position today, suggesting the most likely short-term outcome is a price pullback.

The market internals will receive our close attention and analysis in these early days of the new trend signal.

September 10, 2011. The KP Trend decision models are in a maximum negative position as of this date.

The tactical question for the coming week will be whether the zig-zag sideways pattern that has formed in various US indices over the past month will "resolve" into another leg down in the near term. As we are near the lower extreme of this recent trading range (roughly between 1125 and 1225 on the S&P 500 — we closed Friday at 1154) while our oscillators have moved back to neutral, this bearish outcome is certainly a possibility.

If more weakness does materilaize this coming week, we would expect the more sensitive KP-2 Model to shift from "Minus 2" back to a less aggressive "Minus 1" status sometime during the week. The underlying Trend Model should remain defensive, however.

September 3, 2011. Our trend decision models remain at maximum defensive.

The underlying trend condition (KP-1) is deeply negative, despite the stabilization and rally attempts of the last four weeks, and the recent rally attempts pushed the exposure control study (KP-2) to its maximum defensive condition, which we call "Minus 2" in client communications. Having drawn your attention several times to the persistent relative weakness of the red Model line versus the market index, we would now point out that the Model line appears to be accelerating downward, notwithstanding the recent rally attempts.

With the August month-end last week, we updated several of our longer-term monthly studies, including this one which applies the signals of the KP-1 Model (only) to the S&P 500, using a long/short as well as a long/cash testing protocol. This chart includes only that portion of our Model's history post December 31 2004, or data since the beginning of the Model's out-of-sample / real time history. It seems clear that by dramatically reducing portfolio drawdown (MDD), portfolio performance (IRR) can benefit significantly, to the extent of approximately 500 basis points annually versus buying and holding the index.

August 27, 2011. To review, the KP Model has two components:

1. A slow-moving trend study, labeled KP-1, which changes directions about six times a year and is focused on indentifying intermediate-term trends in the equity markets;
2. An overlay on this trend study whch manages the degree of index exposure within the trend signal condition. This analysis is named KP-2 and suggests exposure adjustments roughly every two weeks;
3. The use of the more active KP-2 overlay is not required. However, for portfolios that are able to make the more frequent adjustments (25 times a year) involved, the use of the KP-2 overlay should result in further improvement in drawdown and long-term returns.

Last week's rally invoked a shift to Minus 2 (maximum hedge / short) by the KP-2 study which, by design, tends to "fade" short-term movements (such as Friday's "Jackson Hole" rally) in anticipation of the natural tendency of financial markets to mean-revert.

Our base trend study (KP-1) remains emphatically negative.

August 21, 2011. The highly volatile stablization attempts of global equity markets continue amid economic and technical market statistics that already qualify as "black swans." And it is not clear that the lowest levels of this decline have been seen. The internal oversold condition last week was mathematically of the same intensity as the October 2008 and June 2010 ("flash crash") events, as can be seen on our oscillator chart.

When markets become this violent, such oscillators are valuable for perspective, but invariably early, as the damaged internal structure takes longer to repair and develop a base.

Last week, the KP-2 study came very close to reverting to maximum defensive (just in time for Thursday's plunge) but it was not to be, so KP-2 remained at Minus 1 / Cash.

The base Trend Model remains solidly defensive.

August 14, 2011. The global equity markets are attempting to stabilize at current levels, and the activity last week — as chaotic and indecisive as it felt — did a great deal to normalize the internals, evidenced by our oscillators.

The base KP Trend Model is "short" and will remain so for the forseeable future, given the very powerful negative impulses that continue to reverberate in the recent calculations. The shorter-term KP-2 overlay has been suggesting "cash" since Friday, July 29 but, given last week's rally attempts, may revert to "Minus 2" (maximum defensive) if/when the markets rally further this coming week (i.e., if prices continue to work off the short-term overextension to the downside).

We will keep you closely appraised of daily developements with the Models.

August 6, 2011. The KP Model signaled "short" at the close on July 27, with the S&P Composite at 1,305 and the Nasdaq 100 at 2,367. Last week, that turned out to be valuable advice.

After Thurday's meltdown, a few comments:

1. In terms of what happened to the markets internally, we certainly experienced a Black Swan event. For example, the NYSE ratio of declining volume to advancing volume for Thursday was 91:1, the most one-sided ratio ever in our 3,875-day database! A lot of sellers were looking for liquidity and there was very little to be had.
2. Conventional interpretation suggests that the market is now deeply oversold, which it is. A good perspective is provided by the chart of the KP 2 oscillator. However, very extreme oscillator readings are often associated with chaotic market conditions and the deeper the oversold, the more time the market generally requires to stabilize and begin a corrective rally process. Deep oversold conditions tend to be bullish but early. It is almost impossible to game the day-to-day dynamics.
3. The KP-2 Model is at "Minus 1" - Cash.

August 1, 2011. Last week was one of the more frustrating periods in the life of the KP Trend Model. However, an indicisive short-term buy-sell-buy-sell sequence is not at all unprecedented, as we have noted with a few historical examples on the updated table of Model Trades.

It is at times like these that we need to keep firmly in mind why we follow such a process: we want to have a very high degree of confidence that our clients' financial ships don't go over the waterfalls. So, every time we hear a distant roar and feel the current in the river accelerating, we go to shore and send scouts ahead to have a look. Meanwhile, the other rafts are passing, shouting at us from the middle of the river, and so on — you get the idea.

For the past six weeks, we have been in an exceptionally news-driven market, but that has also been the case many other times historically. We don't get to make the call as to when a market becomes overly politicized, or when more conventional analysis might apply again.

It does feel that, even when the debt ceiling is finally raised, the underlying long-term trend theme remains one of a topping process similar to the summer of 2007. After all of this back and forth, the long, slow downtrend in the red model line on the KP-1 chart is still intact. This is a personal reflection — it doesn't influence the Model signals.

The Model went officially defensive, or "short," on July 27.

Finally, the more active and nuanced version of the Model — what we call KP-2 — doesn't ever trade against the trend of the base Model, but it is willing to be more active and to take its foot off the pedal once in a while. That process has helped manage the extent of the drawdown from the recent small short-term losses. That algorithm is currently at "Minus 1" — defensive, but in cash rather than short.

July 16, 2011. The values of the Model's components remain very closely balanced around the buy/sell line, which gave rise to a narrow "sell" signal effective the close on Thursday, July 14, only a week after the shift to "buy." This convergence implies that yet another short-term shift is possible (certainly not guaranteed) until the Model "finds" the next meaningful trend event.

Although the annual number of model shifts has risen only slightly in recent years, the spacing between signals is, and remains, highly variable as can be seen from the historical record.

A natural tendency, of course, is to begin to lose confidence in any rules-based process that hits a bump in the road. The historical record, however, strongly suggests staying with the discipline.

July 10, 2011. The KP Trend Model shifted to "Long" on the close of trading Thursday, July 7.

Since the model does extensive filtering and smoothing (all based on the typical "signature" of historical market behavior), the most challenging event is a sudden reversal such as what we've just seen in the last three weeks. Again, the short-term condition, as can be seen for example with the blue line on the KP(2) chart, remains highly overbought and some pause at these levels will be normal. Our long exposure is at "Plus 1" reflecting this statistical likelihood of some near-term corrective action.

The short-term dynamics going forward will be quite important — if the markets can stabilize here in the 1350 area of the S&P 500, and the overbought condition is essentially worn off by the passage of time — then the likelhood of this rally having been a launch event (such as, most recently, post-Labor Day 2010) will greatly improve.

July 3, 2011. A very spirited rally into month-end from the intermediate-term oversold condition has now generated one of the largest overbought conditions in recent years.

The KP Trend Model remains negative, but is adding large positive numbers on a daily basis, such that a sharp reversal will be required this week to keep the internal momentum from generating a "buy" signal in the next few days.

Very extended overbought conditions, especially after deep oversolds, are often associated with launch events — while they still invite some short-term pullback (mean-reversion), more importantly, they often signal the beginning of a longer-term upmove. Certainly not what we were expecting, but that is why we have a Model.

The next few days will be critical.