ETMA_The Evolution

A Trading Edge – What does this mean?

“A trading edge.” This is a widely used, but mostly misunderstood concept in trading. This is the topic for discussion in this week’s Factor Update.
A trading edge is the thin advantage gained over the markets by one’s trading practices. The concept of trading edge is much more pertinent to the futures and forex markets than to the equity markets because futures and forex trading are zero sum games (actually, less than zero sum games because of trading fees). In the equity markets, most everyone gains from a general bull trend except for a small short interest – and short interest players can even be right at the expense of shareholders. In futures and forex there is a short for every long. I can only make money in futures if someone else loses money. Worded a different way, I must be a skilled pick-pocket in order to profit from my trading.

Newcomers to futures and forex trading often falsely believe a large edge can be gained through the selection of the market to trade and the day and price of entry – something I refer to as “trade selection” or “trade identification.” In reality, trade selection offers very little in the way of a trading edge, contrary to popular belief or the ads by the con artists claiming such things as “Right 72% of the time,” or “28% return in just four weeks.”
I trade classical chart configurations, with a focus on:

· Head and shoulders patterns, rectangles and right angled triangles (minimizing such patterns as symmetrical triangles, wedges and trendlines)
· Chart configurations at least 12 or so weeks in duration
· Measured move targets equal to approximately $3,000 per futures contract (or per a forex position with a USD value of $100,000)

Do the chart patterns I attempt to identify give me an edge in trading? Yes, but only a slight edge, and then only to the degree I remain extremely disciplined and patient and apply rigorous risk management protocols.

The real edge in my trading comes in three ways:

1. Getting out of a losing trade quickly and allowing a winning trade to grow in value. In other words, how and when I exit a trade is far more important than how and when I enter a trade.
2. Remaining in trades for days, weeks or even months
3. Increasing my bet size two or three fold on a very small proportion of trades (less than five per year) and being right on a few of the trades I parlay.

I need to comment further on #2 above. My observation is that most novice traders pursue day trading. For the life of me I cannot understand this decision. The HFT operations own the bid/offer spread (often an artificially expanded bid/offer spread) – thereby gaining the most significant edge within a trading day.

This fact places an unusual burden on novice day traders to actually find an edge that works.
To summarize:

· Classical charting principles only offer an edge to the extent patience, discipline and aggressive risk management is used in a trading operation
· The real edge comes from selectively increasing leverage, from jettisoning losers quickly and from holding onto winning trades.

OpenKantu System Generator

Through the past two years at Asirikuy we have developed a substantial amount of knowledge in the generation, evaluation and live trading of automatically generated trading strategies. The first of our software development efforts in this field, Kantu, has now become deprecated within our community (see why below) and I have therefore decided to release this code into the open community in order to avoid wasting all this work and instead encourage others to also explore the possibilities of algorithmic system generation using an already built open framework.
Kantu is a trading system generator that creates trading strategies based on price-action derived rules (comparison of different open/high/low/close values) using OHLC data. Using it you can search for strategies within a selected logic space, finding those that match the statistical characteristics imposed by the user (for example you can search for systems which have a given Sharpe, reward to risk ratio, winning percentage, etc).

Accurate and systematic evaluation of Data Mining Bias (DMB) in trading strategy creation

Hi Bass!

huh, only td i was on this website checking out what tool I can get myself for backtesting surface..Im limited with programming knowledge so everything i get into its relatively hard for me.

Im kinda with MC .NET decision..just need few confirmation on the programming side. Will see what are the prices for programming my strategy and I am ready to translate my strategy to other environment, environment where I can properly backtest multicurrency multiTF strategies.

"This is it. It is now or never. Just got laid off. Now I have to make this trading thing work, once and for all. How would you deal with this and what type of trading strategies would you use in my situation?"

http://www.ftsmodules.com/
http://www.ftsmodules.com/public/mod...micTrading.pdf

Bass, I am hoping this is a hypothetical for discussion purposes??????

Nice link to Excel trading platforms Bass. I am going to spend some time looking through it. :-)

No worries mate, i'm my own boss so no one can lay me off , i just loved the selfie .

Phew..... you had me going for a split second or two. I was thinking...who in their right mind would lay you off? My world was coming apart for a moment!!!

Now you are making fun of me !!!...

Trend Following Wizards – March 2016
April 20th, 2016 · Trend Following, Trend Following Wizards

A lot of red in the March column for the Wizards, with a fairly strong negative average return. The YTD performance is still positive overall with just over a handful of funds being negative too.
Below are the full results as of end March 2016:

Design of a EURUSD Donchian breakout strategy with proven statistical edge

This article appeared first on The Lazy Trader at http://www.the-lazy-trader.com/2012/...-strategy.html

This article is about a Forex strategy that I designed for the EURUSD currency pair. In this exercise I would like to share with fellow Forex traders the concepts and ideas to take into account in order to design a "likely" long term profitable trading system. Steps to follow in the design process, and methods to find the answers to questions such as "Is there statistical evidence to believe this system has a positive edge?" or "How do I choose a reliable parameter set?" or more important yet "How do I know in the future that the system has possibly stopped working and shouldn't be used anymore?"

The basic idea

The initial idea behind the system is to exploit breakouts. And this is not new at all, many successful methods in the past have successfully exploited these inefficiencies where instead of trying to "Buy low - Sell high" we attempt to "Buy high - Sell higher" or "Sell low - Buy to close even lower".

For example, analyzing the daily time frame. If the price of a currency makes a close today that is higher than every close in the previous X days, then we can say that the currency pair is in an uptrend. At that point we could argue that chances are in favor of a continuation and then we could enter a BUY order betting that we will be able to sell higher in the future. The same for a SHORT trade. If the close today is the lowest close of the last X days, we are in a downtrend and we could enter a SELL order hoping to buy back later at lower prices for a profit.

Of course X, which is the number of days we are going to look back for the lowest close and highest close is a number we need to optimize. The greater the value of X, the stronger the trend has been, but presumably the closest the time for such a trend to be over.

For example X = 80 means we would enter a LONG order if the close yesterday is the highest close of the last 80 days, or a SHORT order if the close yesterday was the lowest of the last 80 days.

X = 80 represents a stronger trend, that is more mature, but also probably coming to an end, we don't know.

A smaller value, for example, X = 50 represents a young potential trend that has been formed. it is not so mature, so chances are lower that it is over, but it could also mean a trend that is not solid yet, and could easily be a fake out.

So, both aproaches have a benefit and a drawback. In the first case, chances are higher that it is a solid and strong trend, but it could also be coming to an end soon, so I might get smaller profits. In the second case I risk entering many fakeouts but if I catch one good trend I'm going to be riding it from the begining, or closer to the begining, thus potentially obtaining more profit from it. But the cost of failing many attempts could affect my overall results.

Profitability of the Parameter space

In any case, what is important to understand is that whether X = 50 or X = 80 give positive results in a back test, that doesn't mean that the idea of designing a strategy based on this principle has any merit. For the idea to be truly validated as potentially good, positive results need to be obtained for the neighborhood of surrounding parameters. Let's say X = 80 gives you positive returns on a 10 year backtest, but X = 78, or X = 82 give loses? That means, you probably just curve fitted your system and the system will fail at the slightest market mutation in the future.

That's why we need to verify that surrounding parameters of the space also yield positive results. So, below is a simple Optimization Test from 01/01/2000 to 31/12/2009 run in Metatrader 4, entering trades for 50 day break outs, 55, 60, 65, 70, 75, 80, 85, 90, 95 and 100 day break outs combined with time based exits 6 days to 12 days after the entry.

(Click on image to enlarge)
http://2.bp.blogspot.com/-iWx98Caqd4...ion-Report.PNG

The darker the rectangles the higher the profit obtained in the ten year back-test. White rectangles mean negative profits, that is loses. As you can see all the combinations of parameters yield profits in 10 year back-tests (All green rectangles and no white ones). This means, if you were to choose a given combination and the market mutates, you won't be drastically affected in the future because you have a strong parameter space to withstand market variations. So even though you might not trade with the optimal parameter set in the future you are still bound to have odds in your favor because you are likely trading a parameter set that although is not the optimal anymore, at least it is a profitable one.

I like the combination of 70 day break out with 6 days time based exit. It is not the one that yields the highest profit in the 10 year optimization report, not at all. But it is the one with the best ratio of AverageAnnualReturn/MaximumDrawDown. That's the AAR/MDD ratio. It combines a decent return with one of the smallest Drawdowns in the parameter space and I like that, small draw downs whenever possible.

Positive Mathematical Expectancy of the Entry

We need to make sure that our entry logic is better than a random entry. For that we analyze every single entry in the ten year period and compile the Average Maximum Favorable Excursion (MFE) and the Average Maximum Adverse Excursion (MAE). For more information about what MFE and MAE are, how they are calculated take a look at my article How to measure a system's edge. It is all explained there.

The Mathematical Expectancy Analysis reveals that for the 70 day break out with 6 days time based exit there were 363 entries (363 trades). The Average Maximum Favorable Excursion (MFE) was 160.84 and The Average Maximum Adverse Excursion (MAE) was 121.77. That means, on average trades went in our favor by a distance of 160.84% of the daily ATR and against us by a distance equal to 121.77% of the ATR while we were on the trades (6 days for each trade).

We can see that MFE is greater than MAE, so our entry has a positive edge. Trades went, on average, farther in our favor than against us. This edge can be measured with the Edge Ratio dividing MFE/MAE = 1.32

A truly random entry has an Edge Ratio of 1.00. Our entry has an edge ratio of 1.32 which is a signifficant edge. Our entry logic has a positive edge.

Analyzing our Exit's Edge

But we also need to verify that this is not a lucky occurrence. So, ideally the Edge Ratio is higher than 1.00 not only for the 6 days exit, but for the surroundings. For example the edge ratio should be > 1.00 for 4 days exits and for 8 days exits as well. Back to the calculation again and we obtain:
MFEMAEEdge Ratio
4 day exit125.29100.261.25
6 day exit160.84121.771.32
8 day exit189.41138.551.34

We now have a strong evidence to believe that both our entries and exits have a positive edge and that the parameter space is wide enough to show profitability with different parameter sets. We won't be simply choosing a set of specific unique parameters that happened to yield positive returns in the past.

Setting Targets and Stop Losses with Walk Forward Analysis and Rank Parameter selection

Now, we need to define a Target Profit and Stop Loss for our trades. These values won't be a hard number of pips, because hard number of pips won't be useful when the volatility of the currencies changes. Our distance to Target Point and Stop Loss will be variable and based on currently volatility. That's why we will set them as a percentage of the Average True Range (ATR) at the time of the entry.

We run another optimization where we combine possible Targets from 180% of the ATR to 600% of the ATR combined with Stop Losses from 100% of the ATR to 300% of the ATR. And we cover the parametersspace coarsely, using 20% steps.

(Click on image to enlarge)
http://1.bp.blogspot.com/-kVi9ofswfQ...top-Losses.PNG

We run several optimization reports in two year intervals as follows:

1/1/2000 to 31/12/2001
1/1/2001 to 31/12/2002
1/1/2002 to 31/12/2003
1/1/2003 to 31/12/2004
1/1/2004 to 31/12/2005
1/1/2005 to 31/12/2006
1/1/2006 to 31/12/2007
1/1/2007 to 31/12/2008
1/1/2008 to 31/12/2009

The idea is to obtain the parameter combination of TP and SL that yields the most stable dradown values and the most stable yearly returns. We don't care about the highest return, because there is no guarantee that such parameter set will give the highest return in the future. We want to prioritize stability (Smallest standard deviation of the yearly returns, and drawdowns). We simply want the parameter set with the smallest variability.

We have a bunch of good candidates but the two most stable parameters sets are:
Target 320 % of the ATR, Stop Loss 200 % of the ATR
Target 340 % of the ATR, Stop Loss 200 % of the ATR

The first one is a bit more stable, it has a slightly higher return and a slightly lower maximum draw-down so I'm sticking with that one.

Finally we run a back test of our most stable parameter set for the optimization period 2000 - 2009 and this is what we get:

(Click on image to enlarge)
http://4.bp.blogspot.com/-Pydl4cAVdd...D-Strategy.PNG

Yearly returns

2000 +8.55%
2001 +3.40%
2002 +17.28%
2003 +19.69%
2004 +10.60%
2005 -0.96%
2006 +12.27%
2007 +8.09%
2008 +12.42%
2009 -2.28%

We have obtained a strategy with a positive mathematical expectancy, a wide profitable parameter space and we determined the most stable combination for setting Targets and Stop Losses.

The strategy enters orders with a 70 day break out of the highest close or the lowest close. We set a Target equal to 320% of the ATR which we calculate in terms of pips at the time of our entry based on the value of the ATR at the moment, and we also set a Stop Loss at a distance of 200% of the ATR. This way our strategy is flexible to volatilty changes in the currency pair. Finally if our TP or SL is not hit, we exit the trade 6 days after the entry.

Some additional statistics:

Average Annual Return : 8.90%
Maximum Drawdown : 9.75%
AAR/MDD : 0.91
Ulcer Index : 3.23
Sharpe Ratio : 1.13
Martin Ratio : 2.38


When to stop trading the system?

This is perhaps the most important question because it will give you confidence to stick to the system in the inevitable bad times, and only stop using it when there is evidence that the strategy stopped working. And obviously realize it before your account has suffered a significant draw down.

The fact that the system had a maximum draw down of 9.75% doesn't mean that a higher value won't happen in the future. But we need to decide how much more pain we are willing to suffer before we decide that the system stopped working.

In order to accomplish this we use Monte Carlo simulations. Essentially we distribute all the outcomes of our past trades in different orders. All the losses and winners distributed randomly and not in the particular order in which they took place in the past. We distribute all 363 trades randomly and we do it 100 000 times. That introduces enough randomness in the possible orders or combinations in which the sequence of losses and wins may happen. Then for all 100 000 combinations we analyze final profits and draw downs. We then assume that if our trading in real life reaches the worst Monte Carlo scenarios in terms of profits and draw downs the system has stopped working and it is not safe to use anymore because presumably, the market inefficiency that it exploited is not present anymore.

After analyzing the 100 000 iteration Monte Carlo simulation we obtain the following values:

Worse Case Draw-down 17.63%
Worst case consecutive losses 19

If our system reaches a 17.63% draw-down we can stop using it with evidence suggesting that most likely it has stopped working. As well as if it reaches 19 consecutive losses.

At this point we are done designing our likely long term profitable system. Which we designed and optimized using 2000 - 2009 data. Aren't you curious to know how the system performed during 2010 - 2012?

2010 - 2012 will be our out of sample testing period. Essentially simulating how the system would have traded life from 2010 to 2012 which is a period totally alien and unknown for a system designed using only data from 2000 to 2009.

According to the characteristics of the system, we can expect an average yearly return of 8.90% with a standard deviation of 6.83.

That means there is:
- A 68% chance that our yearly return will be between 2.07% and 15.73%.
- A 95% chance that our yearly return will be between -4.76% and 22.76%

Without further ado here are the results:

2010 +6.12%
2011 -4.27%
2012 +7.83%

Our yearly returns are within the expected intervals. We have one negative year in 2010 but there is no evidence that the system stopped working given that our worst case draw down scenario hasn't happened. The system performs better in 2012 and we obtain a 7.83% return for 2012. Overall a 9.54% equity growth is obtained between 2010 and 2012 with a maximum draw down of 8.51%.

Of course if you want to increase your returns you increase your risk per trade. All the simulations were done with 1% risk per trade, calculating the position size accordingly so that if SL was hit, only 1% of the account was lost. If you increase the risk level to 2% per trade you can expect to double everything, both profits and draw downs. You would have achieved 17.80% average yearly returns in the 2000-2009 period but also a 19.50% maximum draw-down and the worst Monte Carlo scenario would be a 35.26% draw down. What matters in the end is the AAR/MDD ratio (which is always 0.91). It is up to the trader to decide how aggressive he wants to be with his risks per trade levels.

The strategy doesn't use unsound trading principles such as Martingale or Grid trading. It has a positive mathematical expectancy and a wide parameter space of consistently long term profitable parameters. It is not the flashiest or most spectacular strategy in the world, but it is a profitable one nonetheless and a strategy that has been working for the past 12 years, with evidence suggesting that it hasn't stopped working. For sure, I will revisit it at the end of 2013 to analyze its trading results from January 2013 to December 2013.

I hope this article shines some light on the design principles of mechanical Forex systems, or any mechanical system for that matter and that it helps some one out there.

Download source code of the Expert - Lazy Trader Breakout EA

This article appeared first on The Lazy Trader at http://www.the-lazy-trader.com/2012/...-strategy.html

- The simulations have been run in Metatrader 4, with AlpariUK data directly provided by the broker.
- A 2 pip fixed spread was set using the Spread Fixer tool implemented by Asirikuy.com
- The Mathematical Expectancy Analysis was carried out using Asirikuy's Mathematical Expectancy Analysis tool
- The Walk Forward Analysis along with the Parameters Rank Analysis, the Analysis of yearly returns and draw-downs as well as the system's characteristics were obtained using Asirikuy's Draw down Analysis Tool.
- The Monte Carlo simulations were run using Asirikuy's Monte Carlo simulator.

Results - 2014
Results - 2015

My Kind of Music when i'm Trading and Anytime.













https://m.youtube.com/watch?v=RxabLA7UQ9k

Supreme .
https://futures.io/attachments/15131d1276077516

A glimpse of my work, this is the UJ Donchian Channel long term settings 23 Years optimized Data, 18 Years In Sample and 5 Years Out of Sample Data .

Number of Trades is small, but smaller settings will produce more trades at the benefit of Profits, Profit Factor and MAR, E- Ratio is greatly Positive.

https://computeraidedfinance.com/201...-data-sources/

Data Sites with Summaries

scraperwiki.com (The community collects free data from the web and converts it into CSV, etc.)
wikiposit (The author collects free data from the web and converts it into CSV, etc.)
(recommendation) Quandl (Successor of wikiposit: great full-text search, XLS, CSV, JSON export)
Matlab Trendy (The community collects free data from the web and converts it into plots)

Foreign Exchange, Forex, FX

www.forexite.com (Forex, 1 minute ticks), CSV download
ratedata.gaincapital.com (Forex, tick data), CSV download
Forex Trading | Trade Currency Online | Forex Broker | OANDA (Forex, daily), CSV download
TrueFX Tick-By-Tick Real-Time And Historical Market Rates, Clean, Aggregated, Dealer Prices (Forex tick data) after login
www.dukascopy.com (Forex, ticks), CSV
fxtop.com/de/historische-wechselkurse.php?MA=1 (Forex, long histories), Table

Bitcoin

bitcoincharts.com/charts/mtgoxUSD (Daily, many exchanges), Table
www.quandl.com/markets/bitcoin (summary of exchanges)

Debt/Interest Rates/Futures

historical time series
EMMI - European Money Markets Institute | Home (Interest-Rates: Euribor, Eonia), Excel download
RBS Databank (Exchange Rate, Interest Rate, Consumer Price Inflation (CPI), Commodities: Brent Spot, Henry Hub Spot, I P E Brent crude futures 1-pos, I P E Natural Gas Index, OPEC Fixing, US EIA Crude Imports, WTI Daily), Excel download
European Central Bank Interest-Rates and Government Bonds, web tables and partly Excel download
(German) Brse Dsseldorf - Hier wird gehandelt (Bonds, daily), Website table only
(German )Deutsche Bundesbank - Startseite German money and interest-rate statistics
MRCI's Free Historical Futures Prices (many different future contracts, daily snapshots only)

current
(German) Bondboard - Index (Bonds)
(German) www.westlbmarkets.net (2012-07-12: Just went offline, hopefully online again soon), Swaps, Caps, Floors, CSV download
www.boerse-stuttgart.de (web table only) yield curves of bonds per rating and maturity
interest-rate derivatives
(new in the list) erstegroup.com Caps, Floors and Collars in form of Warrents

Stock / Indices

historical time series
finance.yahoo.com/(Stocks, Indices), Excel download
Google Finance: Stock market quotes, news, currency conversions & more, partial Excel download, otherwise website table only
Brsenkurse, Aktien & Finanznachrichten - ARIVA.DE, (Stocks, Indeces), Excel download, German only.
NASDAQ Stock Market - Stock Quotes - Stock Exchange News - NASDAQ.com (Nasdaq), Excel download
mba.tuck.dartmouth.edu (portfolio data), text download
(German) finanzen.handelsblatt.com
Investopedia - Educating the world about finance, Website table only
NASDAQ Stock Market - Stock Quotes - Stock Exchange News - NASDAQ.com, Website table only, intraday, daily
www.standardandpoors.com (S&P 500, daily, yearly)
current
finance.yahoo.com (Stocks, Indices)
Bloomberg Business (Stocks, Indices)

Investment Fonds

Issuer
Union Investment (CSV)

Energy

EU http://www.epexspot.com/de/marktdaten/intraday-handel (German), Website table only
US GDF SUEZ Energy Resources | Commercial Electricity Provider | Historical Pricing Data Login CSV download

General Historical

(German) Brse und Kurse: Aktuelle Brsenkurse in Realtime abfragen - OnVista (Stocks, Indices, Warrants, ETFs , daily Quotes) Website
IndexMundi - Country Facts World-wide econometric data and commodities
Measuring Worth - Relative Worth Calculators and Data Sets US, UK, Japan, China with interest rates, GDP, inflation, gold and silver prices, cool long term, Website table only
Historia.se - Historicalstatistics.org Many countries, especially Sweden, GDP, population and much more, links to other sources
(German) Informationen und Termine zum Oktoberfest in Mnchen | Oktoberfest.de - Die Website zur Wiesn (Price of beer at Octoberfest in Munich, 1817-current)

Economic data

World

IndexMundi - Country Facts World-wide econometric data
OECD.StatExtracts includes data and metadata for OECD countries and selected non-member economies.
Google Data e.g. Market Capitalization in GDP: Google Data on Marcet Cap in % of GDP

European Union

Eurostat
European Central Bank
www.destatis.de Statistisches Bundesamt

United Kingdom

www.statistics.gov.uk

United States

Federal Reserve Economic Data FRED
Census.gov
U.S. Bureau of Labor Statistics
The United States Social Security Administration
U.S. Department of the Treasury
U.S. Securities and Exchange Commission | Homepage
Economagic: Economic Time Series Page (Web, Excel download after login)
Standard & Poor's (Shiller-House Price Index)
Data.gov Public data of all US government organizations

Live Cams

Brse Frankfurt: Aktien Kurse, Charts und Nachrichten (German Dax)

Buy Data:

(I did not try any of these, but they look interesting.)

www.cmegroup.com (derivative prices)
Stock Options Analysis and Trading Tools on I Volatility.com (Implied volatilities)
commodities.euronext.if5.com (Cocoa Future, Robusta Coffee Future, White Sugar Future), 15 min delay, XML
End of Day Stock Quote Data and Historical Stock Prices (End-of-Day Stock), Excel download and more
www.historicaloptiondata.com (Option price data)
Stock Options Analysis & Options Trading Strategy Tools | Livevol (Implied volatilities)

Connors 2-period RSI strategy.

This is a very popular trading method by Larry Connors and Cesar Alvarez. We all know there are no magic indicators but there is an indicator that certainly acted like magic over several decades. What indicator is it? Our reliable RSI indicator.

The modified 2-period RSI trading model makes new highs in 2015.

Over the past few years the standard 2-period trading model as defined in the book, Short Term Trading Strategies That Work, has been in a drawdown. During 2011 the market experienced a sudden and sustained drop which put the trading model into loss. Recall, the trading model had no stops. Since this drop the model has been slowly recovering. Below is an equity graph depicting the trading models equity curve trading the SPX index from 1980. You can easily see the large drop around trade number 135.

http://systemtradersuccess.com/wp-co.../RSI2_2015.png
http://systemtradersuccess.com/wp-co...sults_2015.png

Here is a closeup view of the last 30 trades, which covers about the last seven years. We can see this system is making new equity highs in 2015.
http://systemtradersuccess.com/wp-co...ecent_2015.png
The trading model as originally proposed by Larry Connors is very simple and consist of long-only trades. As a reminder, the rules are as follows:

1. Price must be above its 200-day moving average.
2. Buy on close when cumulative RSI(2) is below 5.
3. Exit when price closes above the 5-day moving average.

All the tests within this article are going to use the following assumptions:

1. Starting Equity: $100,000.
2. Number of shares is normalized based on a 10-day ATR calculation with a $2,000 risk per trade.
3. There are no stops.
4. The P&L of each trade is not reinvested.
5. Commissions & Slippage are not accounted for.

Below is the annual performance of this trading model over the past few years.

http://systemtradersuccess.com/wp-co...turns_2015.png


Modified Trading Model

Back in the year 2013, I explored the robustness of the trading parameters used by the 2-period RSI trading model. That article can be found here. Within that article a slightly modified version of the original trading rules was proposed. In short, they doubled the value of the RSI threshold value (from 5 to 10) and doubled the look-back period for the simple moving average exit rule (from 5 to 10). Finally, a stop value of $2,000 was added. I picked this value because it represents our risk value when scaling the number of shares to trade. Notice we are only risking 2% of our $100,000 account on each trade. Here is a summary of the rule changes.

1. Use a value of 10 as the RSI threshold.
2. Use a 10-period simple moving average as our exit signal.
3. Use a $2,000 hard stop.

Below is a table showing the difference between the original Connors rules and the modified Connors rules.
http://systemtradersuccess.com/wp-co...2_Mod_2015.png
http://systemtradersuccess.com/wp-co...s_mod_2015.png
Our increase in net profit comes at the cost of more trades which is due to the fact of lowering the stand on what we consider a viable pullback. By increasing the RSI threshold from 5 to 10 more setups qualify as a valid entry, we take more trades. But we also increased our look-back period for our exit calculation. Thus, we should be holding some of the trades a little longer in an attempt to make more profit. The stop value does hurt the performance of our model. For example, removing the stop value will result in $178,000 in profit with a profit factor of 2.92. However, were going to keep the stop in place because trading without a stop is something most people will not be doing! It will also help protect us from massive losing trades as seen in 2011.
Other Markets
I tested the 2-period systems on the $SPX.X index which is not a tradable instrument. I did this to get as much backtest history as possible. But how would my modified 2-period RSI system hold up on real ETF and futures markets?

http://systemtradersuccess.com/wp-co...rkets_2015.png
Conclusion

The RSI indicator still appears to be a robust indicator at locating high probability entry points within the major market indices. You can modify the trigger threshold and holding period over a large range of values and still produce positive trading results. I hope this article will give you lots of ideas to explore on your own. Another idea with regards to testing parameters is to independently optimize the parameters over the portfolio of market ETFs instead of using just $SPX. There is no doubt in my mind the RSI indicator can be used as a basis for a profitable trading system.

Get The Book

http://systemtradersuccess.com/wp-co...-That-Work.png

Downloads
RSI2 Strategy Files (TradeStation ELD)
RSI2 WorkSpace (TradeStation TSW)
RSI2 Strategy File (Text file)