PowerStocks Labs have built optimised trend-following timing models for various JSE sectors to allow our subscribers to diversify their investments and time their entry into various diversified markets throughout the business cycle.

Although we have built very robust timing models for the overall ALSH index, it is a known fact that various JSE sub-sectors may be in differing bull/bear phases to that of the ALSH index. More recently we saw one of the strongest bull runs in the JSE's history, and yet the Construction & Materials index declined a whopping 21%. Similarly, it was the retail and industrial counters that came alight in the more recent bullish phase while some other sectors bumbled along.

The impact of the economic cycle on individual shares' earnings vary at different times depending which sector they are in. This in turn, dictates when the JSE will rally for those counters in anticipation of improved earnings. With our trend-following timing models for JSE sectors, investors and traders can time their entries and exits into various JSE sectors.

We have launched this service with 6 sectors we believe display differing synchronicity with the JSE index. This means that in addition to playing the very robust signals we give for the overall ALSH index (by playing TOP-40 and RESI-20 instruments), our subscribers may at differing times be getting signals from other sectors that are not 100% in synch with the overall JSE index. This ensures signals will be present for our subscribers to play regardless of whether the JSE is in a bull or bear market. The signal diversification ensures that if one signal leads to a small loss it can be covered by large gains on a trade made on another sector and we do  not lose time waiting for another signal on the sector that just generated the small loss.

The concept we provide is that PowerStocks subscribers can play reliable timing signals on various diversified markets, many of which are readily tradeable on the JSE and your online brokerage via simple, cheap Exchange Traded Funds (ETF's). More advanced or adventurous traders can play these signals with individual large-cap shares within the target sectors to boost their returns even further. Whilst our backtests of performance excluded leverage (our policy), advanced subscribers could super-charge returns with leveraged instruments of the ETF's or shares in question.

The six diversified sectors we have chosen to build additional customised trend-following timing models for are :

1. Gold in ZAR/ounce, which can be traded via the NEWGLD ETF
2. Industrials, which can be traded via the SATRIX-INDI ETF
3. Financials, which can be traded via the SATRIX-FINI ETF
4. Listed Property, which can be traded via the PROPTRAX ETF
5. Construction and Materials which can be traded via MUR, BSR, AEG and GRF
6. General Retail which can be traded via WHL, FOS, MSM, TRU, MPC

These sectors can be played in addition to our highly robust signals for the overall ALSH index. As we have shown in previous research, the Resources-20 sector is highly correlated to the ALSH index and works exceedingly well with our ALSH timing models. So including the TOP-40 and the Resources-20 index, PowerStocks subscribers now have access to 8 diversified markets at any point in time from a market timing perspective.

These sectors can be easily traded with exchange traded funds (ETF's) as a single share purchase to give you immediate access to all underlying shares in a sector without having to purchase the individual shares themselves. For the TOP-40 there is the SATRIX-40 and for Resources, there is the SATRIX-RESI. We mention the names of some ETF products but in reality, any ETF that tracks a particular sector will suffice. You can elect to trade the ETF straight with no gearing (vanilla) or with gearing through SSF's and CFD's. Standard Bank online share trading provide CFD's and SSF's for all the SATRIX ETF variants.

The last two sectors, (construction and retail) in addition to providing differing synchronicty to the overall JSE index, also provide for very powerful JSE-beating returns when timed properly. However they are not tracked by any ETF product. In reality you could trade the 1 or 2 largest shares (by market-cap) in these sectors and have an instrument that will track the sector by at least 98% correlation but we advise at least 2-3 shares to capture the full essence of the sector and lower your risk. We give guidance on each of these sectors with a list of 5 shares that dominate the sector and track it with at least 98% correlation. This allows you to "build your own mini-ETF" without having to incur the cost of purchasing too many shares, to capture the gains of that sector.

The table below depicts the 10-year correlation between the 8 JSE sectors we now run timing models for and the overall JSE index.

We see that the TOP-40 (TOPI) and Resources (RESI-20) sectors are very closely correlated to the overall ALSH index and will thus work very well with our breadth-based ALSH timing systems such as HedgeTrader etc. We note that GOLD and particularly the listed property index, have much lower correlations and will thus have very different synchronicity with the overall stock market. This makes them excellent diversifiers to your portfolios. It is thus feasible to envision that while our timing models for the JSE have you OUT the market, that our timing models for RETAILERS, GOLD or LISTED PROPERTY may very well have you IN the market!

As a subscriber, you will use the new timing models various ways. The first is to use them to initiate your entries into various sectors as they ebb and flow during the economic cycles. This ensures that you are kept abreast of the highest performing sectors at all times to capture their gains.

The second use is to validate the purchase of a particular share you have your eye on. Let us assume you run one of our strategy screens and come up with MUR as a candidate. You can then refer to the CONSTRUCTION timing model to ensure it is "IN BULLISH MODE" before committing your share purchase. After all, it would be silly to purchase MUR if the construction sector is on a downward slide! As we saw recently, one might have seen the ALSH is in bullish mode and went ahead and bought MUR and suffered a 21% loss! The sector timing models ensure you only buy shares when their relevant sectors are trending upwards - don't fight the trend!

You could also use the timing models to bring diversification to your portfolio. Certainly if GOLD and PROPERTY or RETAIL enter bullish phases and get BUY signals, it would make sense for you to load up on shares (or an ETF) representative of these sectors to bring diversification to your portfolio since they have lower correlations with the JSE.

After extensive research into quantitative price-based Trend-Following theory we have built sophisticated (non-breadth) timing models tailored to each of the sectors using 12-14 years of daily index data. Each sector or index has its own liquidity, volatility and cyclicality to the economic cycle and therefore we went against a "one size fits all" algorithm to time each respective sector.

Traders and investors who subscribe to a trend-following strategy do not aim to forecast or predict specific price levels or concern themselves with fundamentals - they simply jump on the trend and ride it as long as possible (until it bends).

Trend-following is not as powerful as our breadth-based models for the ALSH index, since it relies on price action alone, but it still produces amazing results regardless. Remember with these models, some up-tick is required before the model determines a new up-trend is in progress and thus the models are not good for picking troughs, although when one looks at the whole signal beginning to end it would appear that the system would have got in close to the trough. Similarly, a reasonable "pull-back" is required for these models to determine that a new "downward" phase of the index or sector is in progress. As such it gets you out AFTER the main peak and some profits have to be sacrificed in the process.

As a result, trend-following systems get you in after the trough and after the peak and for best returns, you are encouraged to get in as early as possible with the signals, since it is possible if you get in too late that the gains you make are exceeded by the "sacrifice" required to determine the exit.

Trend-following models do not produce the same win-rates that our subscribers have become spoilt with on our breadth-based systems for the overall JSE ALSH index. They are sometimes characterised by many small losses in a row (false starts) before eventually latching onto the main trend. If the target market is range-bound (i.e. not trending up or down) it can lead to whipsaws. However the systems we have calibrated ensure a reasonable win-rate to ensure you do not become discouraged after a string of losses. Small losses that are cut short are common with these models, but once you have latched onto a trend, you stick with it for huge gains that swamp the small losses. It can be best described as a "cut your losers early and let your winners run." strategy. The gain/loss ratio thus tells the real story and not the win rates (percentages of trades resulting in a profit.)

The strategies sit tight when a strong trend is in progress, meaning you can be long for up to 18 months at a time! But it can also kick you out after merely weeks. It is thus difficult to describe trend following as either short, medium or long term - it is all of them - dictated by the trend and price action!

Quite often with trend following, the system could tell you to sell your portfolio after a bit of a downturn and a week later tell you to buy back in again. Quite often you could land up selling in the trough and buying again a bit after the trough. You may be forgiven for thinking the system is inaccurate but this phenomenon is a characteristic of the safety devices built within the models to keep you out of extended or even brutal rapid bear markets. As long as you got into the trend in the beginning, even when you sell in the trough, you will be banking a nice profit. And generally when the system tells you to get back in again, another nice up-trend is on the way. This highlights the importance of getting onto these signals as early as possible.

Money and risk management is an important (no, crucial!)  part of trend-following based investing/trading. It is critical if you wish to be successful with us and make lots of money over time that you exercise proper money and risk management. We urge you to read this section carefully - it is normally the "ah ha" moment for many beginners who go on to become highly successful at trading.

We have built stop-losses into most of the sector timing models. There are initial stop-losses and trailing stop losses. The values used for the initial stops are dependant on the volatility of the index/sector at the time of opening the trade. Your initial stop is used to calculate your position-size based on a fixed risk. If the initial stop is x% below the trade open price then the investor can calculate the maximum risk for the trade and configure the trade with a properly sized bet. Generally speaking when a market is highly volatile, the risk is higher and the initial stop must be widened to ensure you are not stopped out prematurely.

Each timing chart for each sector also displays a trailing stop loss once the share/index has exceeded the initial stop loss by a certain amount. This serves to lock-in profits from being eroded by declines. The value of the trailing stop loss is computed automatically by our algorithms depending on the volatility of the share/index in question, to ensure you are not prematurely stopped-out. When the trailing stop-loss exceeds the initial stop loss, then the trailing stop loss takes over the governance of whether we exit a trade or not and we forget about the initial stop loss.

The combination of an appropriate initial-stop (to size your bet appropriately to risk) combined with a trailing stop (to lock in profit), we call our "Smart Stop" system. It is demonstrated below, with the yellow line forming the initial-stop and the red line the trailing stop.

When we notify subscribers of a new open trade, we will give them the percentage to use for the initial stop. They can then use this to configure an appropriately sized trade either manually or with our PowerStocks Trade Risk Configurator (PTRC) tool.

As a general rule we do not recommend any one trade exceeding 2-5% of your trading capital (which in turn should not exceed 40% of all your JSE funds). If you know the risk via a % we give you as an initial stop then you calculate you trade size as follows:


Let's assume you have R100K trade capital and get a signal that GOLD is now a BUY. Your maximum risk is say 2% per trade which is R2,000. This is the maximum we are willing to lose on this trade. Let us assume in our alert we also tell you that Smart Stop is advising a 5% initial stop on the NEWGLD ETF. You take the R2,000 and divide it by 5% to get a maximum exposure of R40,000. You place your trade for R40,000 of the GLD ETF and set a stop at 5%. At the worst you will hit your stop and lose the whole R2,000. You will have limited your risk (total loss) to 2% of your trading capital. You have sized your trade appropriately, to the shares volatility, to limit your total losses. No single trade can wipe you out. Even a string of losses cannot wipe you out.

The larger the current volatility of the target index, the higher the initial stop will be. The higher the initial stop the higher your risk. By plugging in a higher initial stop into the MAX EXPOSURE formula above, you will notice that it will let you place FEWER funds into the market. In this fashion you place more money into the market in less volatile (risky) periods and less money during more volatile (risky) periods. If GOLD was really bouncing around but we established a trend was in progress then maybe we would issue a SmartStop of 10% instead of the 5% in the previous example. Plugging this into the formula will yield a value of MAX EXPOSURE = (R100,000*2%)/10% = R20,000. So in a more volatile and risky market the money management rules would halve your trade size from R40,000 down to R20,000.

With our PTRC tool, you can input gearing, slippage, brokerage and even the timing signal you are acting on to properly configure the trades risk. Our timing signals have various degrees of statistical confidence which we use to configure an appropriate trade risk percentage from 2-9%. Our initial hypothetical GOLD trade discussed above, where we issued a 5% initial stop is configured below.

The trade size is  less than the R40,000 we discussed previously because of brokerage and slippage. PTRC tells you how many shares to buy and what to set your initial stop loss to. As you notice, it allows for gearing as well if you plan to use CFD's or SSF instruments. Just e-mail research@powerstocks.co.za if you would like a copy of PTRC (subscribers only.)

The table  below shows the performance of the various trend-following timing models for the respective sectors. Depending on the amount of historical index data we have test periods range from 5-14 years:

System Gain is the total return after all timing trades with the proceeds of the previous trade fully vested into the following trade. It excludes interest on cash for non-vested periods when the strategy is sitting in cash. Sector Gain is the buy-and-hold gain of the sector during the period in question. Vested CAGR is the compound growth rate achieved by the timing model during vested periods only. It is a gauge of the "earnings power" of the strategy when it is vested in the JSE. You can see that the trend following timing model for the PROPTRAX ETF produced the most bang for your buck, followed by CONSTRUCTION and then RETAIL.

A historical sample of the timing model for the Construction & Materials sector is shown below:

If you followed the above system, shoving your money under a mattress every time the system gave a SELL signal and re-investing all proceeds of previous trade in the next BUY signal then your total return on an initial R1.00 investment is shown below. You would have grown your R1 to just under R7, versus the buy-and hold strategy which would have grown the R1 to R4. But more importantly, you would have done this by being exposed to the JSE only 53% of the time! If we include modest interest earned during non-vested periods, the returns would be above R10.

The CONSM index above includes PPC. If one excluded PPC and just invested in MUR, BSR, GRF and AEG, your returns would have been R15 in the above chart! These are the 2nd-5th largest shares in this sector but all of them comprehensively outperformed the sector index itself. While on this topic, all of the above shares track the CONSM with a 98% correlation or more.

There is a new sheet in the JBAR report called "SECTOR TIMERS" that includes the timing signals for the 6 sectors we now track.

Some outdated samples are below:

Notice how in the above chart for RETAIL that we were forced out after  a 23% profit, in the midst of a large pull-back, and then a few days later got another signal to get back in again - which led to another 19% gain. This may seem ineffective but the system determined during the pull-back between the two signals that risks were high we were possibly entering a larger correction and pulled the sell trigger. No harm was done, it merely forced us to lock in healthy profits and on many occasions in the 12 year past this action would have saved us from a brutal correction. In any event the system got us back in again to capture more gains. This is just something you will have to get used to and trust with our trend-following systems.

Traders learn to just follow the signals blindly no matter how ridiculous they may seem. However, PowerStocks subscribers would do well to compare the trend-following sector signals with the SuperModel and LBYC investment timing models to make educated decisions about liquidating though. In the above case we would probably have ignored the SELL since both SuperModel and LBYC were showing such bullish signals at the time.

As we mentioned there are no ETF's for CONSM and RETAIL but you can easily create your own mini-ETF. In fact you can create your own mini-ETF for any of the sectors above. Just choose 3-4 of the largest cap shares in each sector and you will be within a 98% correlation to the sector.

Doing this often produces far superior returns to the sector itself. For example, by sticking to combinations of MUR, BSR, GRF and AEG for the construction sector and sticking to combinations of WHL, FOS, MSM, TRU and MPC for the retail sector you are likely to achieve returns far in excess of the sector index as these are large dominant growth shares in these respective indices.

In addition our 10-year tests have shown them all to be over 98% correlated to the respective index they belong to. Note that a high correlation to a sector does not mean closely matching returns. WHL out-performed the RETAIL sector by over 3x in last 10 years but is 99% correlated to its movements. It just means when RETAIL trends up, so does WHL and vice versa.

Even if you just pick one of the shares mentioned above chances are very good you will tap into the underlying sector trend and outperform the sector growth.
We do not advise you use shares other than those we mention as results may be unpredictable. More shares is better from a diversification point of view so although you can pick one share we recommend at least two.

If you truly want to mirror the CONSM index returns you need to include PPC in your portfolio. However as we stated this has the effect of reducing performance somewhat since PPC is not a growth stock. Your result will be closer to the sector index though. In fact just PPC and MUR combined in equal amounts produce an index that is 99% correlated to the CONSM index movements (trends) and matches its returns in excess of 85%. We do not advise using PPC on its own though. Excluding PPC from your "mini ETF" portfolio will achieve returns far higher than that of the construction sector.
Make a Free Website with Yola.