Jump to content

PositiveDev's trading journey


Recommended Posts

It's really been far too long since I last posted. I have been doing more analysis and reached a point where I realised I need more tools so I've been working through some online courses on statistics and probability; https://www.thegreatcourses.co.uk/, run by an award winning professor, Professor Starbird from the University of Texas.

 

I'm looking to bring an added dimension to my analysis work, with a view to returning to trading when I have evidence of a tradeable edge.

Link to comment
Share on other sites

  • Replies 1.1k
  • Created
  • Last Reply

One of the courses I'm working through is Effective Thinking through Mathematics, and although it's obviously about mathematics, it's really about how you can take existing ideas and develop them further to generate new insights. I'm finding it quite fascinating because I wasn't particularly enamoured with maths back at school, and yet the way it's being taught in this course makes it very interesting.

 

One example is very simple, and that it the difference of squares, so the difference in the area of two squares. The area of a square is one side squared so;

 

Area = x2

 

If one square is x and another is y then the difference is;

 

x2 - y2 or (x-y)(x+y)

 

This course teaches you to visualise what is actually happening which then informs an understanding of the equation. From there I was interested in how you would calculate the difference of cubes, and I worked out the equation for the difference of cubes as being;

 

x3 - y3 = ((x-y)(x+y))y+x2

 

The above works, albeit you need to choose the higher of your two valuations for x - it's not the most user friendly way you can create the formula.

 

This course has caused me to start questioning my own beliefs about the market, what's possible and what isn't, the chief belief is that markets are not random.

 

In order to test this I created program that would generate random charts, here are some of the results;

 

Randomstock3_zpse3e75a20.png

 

Randomstock2_zpsa291f554.png

 

[/url]

 

Capture_zps09c06776.png

 

 

Most of these do indeed look quite a lot like stock or index charts - although there seems to be more structure to the bull and bear phases. Markets may have characteristics of randomness but I don't think it's accurate to say they are completely random, certainly not in the long term at any rate.

 

 

Another aspect of the course is how to solve problems, principals to work through maths problems and otherwise, and this has recently led my analysis work in a different direction.

 

What is the highest probability trade you can take on the Dow Jones future market in any given day, for a trade closed the same day? I found myself asking this question, and I wanted the answer. Investigating the data covering the last year enabled me find the most optimum ratio between risk and reward over a given day, and to my great surprise, it was the same values that I had already used previously in my trading - a co-incidence perhaps. More than that, the data shows that it is simply not worthwhile at all to short for trades closed before the close, relative to the increased probability of going long, although that aspect needs work on more years data, since some years tend to be more bullish than others. Furthermore, the probabilities of a successful trade vary significantly over the duration of any given trading day, not that this should be news to anyone, but quantifying these probabilities brings it to new life. So there are commonalities across all days, that repeat. Interestingly I'm letting the data push me to a conclusion, not the other way around (I previously tried to crowbar strategies onto the market without conducting adequate research. I've been wanting to do work of this sort for some time but until now couldn't quite manage to grasp the complexities involved in it.

 

 

This is completely different work to that of the Pearson correlation analysis I've been working on, so perhaps I can pull this work together to create something.

Link to comment
Share on other sites

Investigating the data covering the last year enabled me find the most optimum ratio between risk and reward over a given day, and to my great surprise, it was the same values that I had already used previously in my trading - a co-incidence perhaps

 

And the great problem with this is, as I'm discovering through my analysis is that whilst it is the optimum taken over the day as a whole, the probabilities develop and change over the course of the day so to use a one size fits all is the completely wrong approach.

 

What my very recent analysis appears to be showing me is that target and stop must be dictated by market context, and that there is a random element to price action in concert with causal elements - it's fascinating to think about and study.

 

My analysis is also showing up correlation to be a very blunt instrument, a clumsy way to try to define descriptive relationships (for that is what correlation is) within the market.

 

It's fair to say that taking these online courses has helped show me what a proper scientific study and process looks like and keying off of that with my recent work.

 

I need to broader the testing to see if these apparent insights I have translate to other markets, other types of instruments besides futures (I've only looked at Dow futures so far).

 

I'm starting to get a feel for how the next strategy is going to look.

Link to comment
Share on other sites

  • 2 weeks later...

I think some of the work I've being doing of late is the best work I've ever done. It's really opened my eyes in terms of understanding what an effective intraday strategy looks like, how difficult it is to create an effective strategy without doing considerable data-mining work, and why the vast majority of intra-day traders lose.

 

The recent work centers on identifying the highest probability opportunities to trade and trading them. It is entirely data driven so trade entries will only occur when all the variables line up.

 

I had previously developed a proprietary indicator and developed a strategy around that, and whilst the strategy worked for a number of months it the ceased. The indicator is very highly correlated with Dow Jones futures, a Pearson co-efficient of 0.8. However a correlation co-efficient is only a description of a relationship, it does not say anything about whether one variable (the indicator value) causes the other (Dow Jones futures) to any degree. My research suggests that there is a degree of causation from one variable to the other, to the extent, and I can use that to factor into a trading strategy to push the probabilities further in the right direction.

Link to comment
Share on other sites

I think some of the work I've being doing of late is the best work I've ever done. It's really opened my eyes in terms of understanding what an effective intraday strategy looks like, how difficult it is to create an effective strategy without doing considerable data-mining work, and why the vast majority of intra-day traders lose.

Interesting.

 

So is that final comment base on proven research? Also is that intra-day traders losing based on some form of Q.A or charting? Including those that trade purely on news? Or combination of everything thrown in?

 

Just curious, as I rarely trade this way, largely exclusively on flow when I do, with the strong belief it's generally a bad idea for all but the most talented, as like you say, I suspect most did in fact lose.

Link to comment
Share on other sites

Interesting.

 

So is that final comment base on proven research? Also is that intra-day traders losing based on some form of Q.A or charting? Including those that trade purely on news? Or combination of everything thrown in?

 

Just curious, as I rarely trade this way, largely exclusively on flow when I do, with the strong belief it's generally a bad idea for all but the most talented, as like you say, I suspect most did in fact lose.

 

From Walsh Agency Inc. Conneticut

 

"We surveyed more than a thousand experienced futures brokers and asked what, in their experience, caused most futures traders to lose money. These account executives represent the trading experience of more than 20,000 futures traders. In addition, most of these Account Executives (AEs) have also traded or are currently trading for themselves. Their answers are not summarized because different traders make (and lose) money for different reasons. Perhaps you may recognize some of your strengths and weaknesses. Yet, many of the reasons given are very similar from broker to broker and client to client. The repetitions stand to demonstrate that, alas, many futures traders lose money for many of the same reasons. Perhaps these statements from experienced brokers can make a contribution to you, who make this sometimes fickle, often intricate, always interesting marketplace of futures trading possible. Here is what they said:

 

Many futures traders trade without a plan. They do not define specific risk and profit objectives before trading. Even if they establish a plan, they "second guess" it and don't stick to it, particularly if the trade is a loss. Consequently, they overtrade and use their equity to the limit (are undercapitalized), which puts them in a squeeze and forces them to liquidate positions.

Usually they liquidate the good trades and keep the bad ones. Many traders don't realize the news they hear and read has, in many cases, already been discounted by the market.

After several profitable trades, many speculators become wild and unconservative. They base their trades on hunches and long shots, rather than sound fundamental and technical reasoning, or put their money into one deal that "can't fail."

Traders often try to carry too big a position with too little capital, and trade too frequently for the size of the account.

Some traders try to "beat the market" by day-trading, nervous scalping, and getting greedy.

They fail to pre-define risk, add to a losing position, and fail to use stops.

They frequently have a directional bias; for example, always wanting to be long.

Lack of experience in the market causes many traders to become emotionally and/or financially committed to one trade, and unwilling or unable to take a loss. They may be unable to admit they have made a mistake, or they look at the market in too short a timeframe.

They overtrade.

Many traders can't (or don't) take the small losses. They often stick with a loser until it really hurts, then take the loss. This is an undisciplined approach...a trader needs to develop and stick with a system.

Many traders get a fundamental case and hang onto it, even after the market technically turns. Only believe fundamentals as long as the technical signals follow. Both must agree.

Many traders break a cardinal rule: "Cut losses short. Let profits run."

Many people trade with their hearts instead of their heads. For some traders, adversity (or success) distorts judgment. That’s why they should have a plan first, and stick to it.

Often traders have bad timing, and not enough capital to survive the shake out.

Too many traders perceive futures markets as an intuitive arena. The inability to distinguish between price fluctuations which reflect a fundamental change and those which represent an interim change often causes losses.

Not following a disciplined trading program leads to accepting large losses and small profits. Many traders do not define offensive and defensive plans when an initial position is taken.

Emotion makes many traders hold a loser too long. Many traders don't discipline themselves to take small losses and big gains.

Too many traders are underfinanced, and get washed out at the extremes.

Greed causes some traders to allow profits to dwindle into losses while hoping for larger profits. This is really lack of discipline. Also, having too many trades on at one time and overtrading for the amount of capital involved can stem from greed.

Trying to trade inactive markets is dangerous.

Taking too big a risk with too little profit potential is a sure way to losses.

Many traders lose by not taking losses in proportion to the size of their accounts.

Often, traders do not recognize the difference between trading markets and trending markets.

 

Lack of discipline is a major shortcoming.

 

Lack of discipline includes several lesser items; i.e., impatience, need for action, etc. Also, many traders are unable to take a loss and do it quickly.

Trading against the trend, especially without reasonable stops, and insufficient capital to trade with and/or improper money management are major causes of large losses in the futures markets; however, a large capital base alone does not guarantee success.

Overtrading is dangerous, and often stems from lack of planning.

Trading very speculative commodities is a frequent mistake.

There is a striking inability to stay with winners. Most traders are too willing to take small profits and, therefore, miss out on big profits. Another problem is undercapitalization; small accounts can't diversify, and can't use valid stops.

Some traders are on an ego trip and won't take advice from another person; any trade must be their idea.

Many traders have the habit of not cutting losses fast, and getting out of winners too soon. It sounds simple, but it takes discipline to trade correctly. This is hard whether you're losing or winning.

 

Many traders overtrade their accounts.

 

Futures traders tend to have no discipline, no plan, and no patience. They overtrade and can't wait for the right opportunity. Instead, they seem compelled to trade every rumor.

Staying with a losing position, because a trader's information (or worse yet, intuition) indicates the deteriorating market is only a temporary situation, can lead to large losses.

Lack of risk capital in the market means inadequate capital for diversification and staying power in the market.

Some speculators don't have the temperament to accept small losses in a trade, or the patience to let winners ride.

Greed, as evidenced by trying to pick tops or bottoms, is a frequent error.

Not having a trading plan results in a lack of money management. Then, when too much ego gets involved, the result is emotional trading.

Frequently, traders judge markets on the local situation only, rather than taking the worldwide situation into account.

Speculators allow emotions to overcome intelligence when markets are going for them or against them. They do not have a plan and follow it. A good plan must include defense points (stops).

Some traders are not willing to believe price action, and thus trade contrary to the trend.

Many speculators trade only one commodity.

Getting out of a rallying commodity too quickly, or holding losers too long results in losses.

Trading against the trend is a common mistake. This may result from overtrading, too many day-trades, and undercapitalization, accentuated by failure to use a money management approach to trading futures.

Often, traders jump into a market based on a story in the morning paper; the market many times has already discounted the information.

Lack of self-discipline on the part of the trader and/or broker creates losses.

 

Futures traders tend to do inadequate research.

 

Traders don't clearly identify and then adhere to risk parameters; i.e., stops.

Most traders overtrade without doing enough research. They take too many positions with too little information. They do a lot of day-trading for which they are undermargined; thus, they are unable to accept small losses.

Many speculators use "conventional wisdom" which is either "local," or "old news" to the market. They take small profits, not riding gains as they should, and tend to stay with losing positions. Most traders do not spend enough time and effort analyzing the market, and/or analyzing their own emotional make-up.

Too many traders do not apply money management techniques. They have no discipline, no plan. Many also overstay when the market goes against them, and won't limit their losses.

Many traders are undercapitalized. They trade positions too large, relative to their available capital. They are not flexible enough to change their minds or opinions when the trend is clearly against them. They don't have a good battle plan and the courage to stick to it.

Don't make trading decisions based on inside information. It's illegal, and besides, it's usually wrong."

 

Also see - http://travismorien.com/FAQ/trading/futradersuccess.htm

Link to comment
Share on other sites

Interesting.

 

So is that final comment base on proven research? Also is that intra-day traders losing based on some form of Q.A or charting? Including those that trade purely on news? Or combination of everything thrown in?

 

Just curious, as I rarely trade this way, largely exclusively on flow when I do, with the strong belief it's generally a bad idea for all but the most talented, as like you say, I suspect most did in fact lose.

 

I wasn't clear on whether you meant research I'd encountered or my own personal research. The post above refers to a Walsh Agency Inc. survey. My own recent research shows that short term price movements in stock index futures are characteristic of random price movement. That is the structure of price moves bears similarity to the the types of moves expected from a random price chart, as I have shown in previous posts. It's very important however to distinguish that whilst stock index futures prices have characteristics of random price movement on the short term, they are not completely random. They are random enough to cause traders with risk/reward strategies that are too skewed one way or the other to lose more than they win, but not random enough to prevent yield being generated if the the right strategy is chosen. In a truly random scenario I do not believe it would be possible to generate yield over the short term.

 

My research shows that stock index futures are short term random characteristic and what I call long term mechanistic (purely supply/demand driven). There is a swing from one to the other during the transition from the short to the long term. An interesting point to consider is that some short term traders might employ a risk reward strategy of perhaps 1:2 or 1:3. This type of strategy is what I might diplomatically call exceptionally ambitious given that the most talented money managers that have ever lived have been able to consistently pull annual yields of perhaps 80% or 90% for a few years, and over longer terms perhaps pulled annual yields of 30% - 40% - these are the best of the best hedge fund managers I'm referring to, so I'm creating a euphemism when I say "exceptionally ambitious". It's well known that it is considerably more difficult to trade successfully over the short term so those traders looking for 200%/300% on one trade - consistently - are indeed "exceptionally ambitious", to their detriment.

 

I've been working on a new strategy (Strategy III) recently to take advantage of the insights from my recent research and it looks interesting so far - it's very counter-intuitive in many ways - my first instinct is to see if I can rip it to shreds, search for the flaws within it, before I employ it in the market. I'm still going through that process but it does look interesting at this point.

 

This chart shows performance per contract on Dow Jones futures (YM) from 2nd January to 9th December 2014. $3880 per contract gross yield over that period (chart is net of $4 per trade commission so actual yield of $2992 per contract). This performance is equivalent to a 17.4% gain on capital risked. I'm currently going through a process of backtesting to see how it performed over preceeding years, and on other stock index future products. On a reasonable bank of $10000 per contract traded the maximum drawdown is around 5% (during 2014 at least) which is good for what appears to be a consistent and stable performance, but I need to do more testing over other years and to see how correlated it is with general market performance, both during bull and bear markets.

 

Stategy III 2014

StrategyIII_zps068ed551.png

Link to comment
Share on other sites

Thank you for the explanation, I understand a little better now, Sorry I didn't explain myself clearly but I'd basically missed the point about it being Walsh Agency Inc research.

 

This sort of trading is not for me but I do find it fascinating and like to read your progress, hope it all comes together for you.

Link to comment
Share on other sites

  • 1 month later...

Based on all of the recent research I've done, I've created a new strategy, and backtested it over the last 6 years to see how it performed;

 

 

I've charted the performance of the strategy against the performance of the underlying index - in this case Dow Jones futures;

 

2010 was the only non-profitable year, I will need to take a closer look at that to see why that was the case, but even so the loss on capital that year was largely a reflection of the cost of trading (commissions etc) that I have factored into these figures. The Win Loss ratio of 0.97 that year highlights that.

 

 

2009

20150202%20YM%200851%202009%201.84%2074_

 

2010

20150202%20YM%200851%202010%200.97%2074_

 

2011

20150131%20YM%200851%202011%201.52%2069%

 

2012

20150131%20YM%200851%202012%201.61%2060_

 

2013

20150131%20YM%200851%202013%201.63%2060_

 

2014

20150131%20YM%200851%202014%201.92%2051_

 

Gains_zpsfeabeenj.png

 

These are good results, showing the per contract gains for Dow Jones futures - the performance is broadly similar year to year, it's very selective, averaging 5.4 trades per month. The return on risk figures is % gained over the total capital risked across the year. The win/loss ratios are fairly consistent.

 

I would be very interested to see how this performs on S&P 500 futures (ultra competitive market).

Link to comment
Share on other sites

  • 8 months later...

Well I am back. I cannot believe my last post was February. Finding time to work on trading is challenging with a 19 month old and a full time salaried role. I'm confined to the small hours when everyone is asleep.

I was in a very difficult place with my trading research back in February. Essentially I had coded my trading strategy into an Excel spreadsheet, which ended up being over 600,000 rows of data with formulas across god knows how many (20+?) columns, this required significant processing power. So much so I had to buy a much faster PC. Despite buying a very fast PC it still took around 10-30 seconds to process any change in the spreadsheet. This meant it was an incredibly painful experience to analyse the performance of the strategy, performance charts had to be created manually - it was awful. Additionally excel is notorious in terms of being a hotbed for unseen errors and it was certainly not designed to do what I was forcing it to do.

Anyway fastforward to now. In the last few days it's become obvious that I needed to pursue a new path and go down the programming route. I use Multicharts and it has a built in programming language that allows you to code trading strategies, its a variant of the same one within Tradestation (well known within the industry). It has only taken me a few hours over the last 3 days to code my strategy and have it run natively within Multicharts. It wasn't as difficult as I'd originally thought. This now means I can analyse my strategy very quickly with various stats and charts available. I can make changes to the strategy in literally seconds and check on how differing variants perform - something that was previously painful, time consuming and laborious to do is now almost effortless. Moreover when I do trade again, it will be automated with my trading platform running my automated strategy program, entering and exiting trades according to my strategy rules. I will be able to run different iterations of the same strategy on many other futures markets very quickly so I can readily explore other markets for trading using the basic principles inherent within this strategy.

My ultimate goal is to create a robust trading strategy that is consistent over a multi-year basis, and one that is readily scalable. The one I am working on that I have just programmed is surprisingly simple yet the backtested results show that it is effective - it is a variant of the one described in the last post.

Here are some numbers that reflect backtesting results using the strategy to trade Dow Jones futures from 2010 to mid 2015 (this is a long only strategy);

Opening account value is $100,000
10 contracts bought per trade

Annual%20Period%20Analysis_zpsgzcgxtvp.p

Equity curve

Equity%20curve%20close%20to%20close_zps9

Buy & Hold returns over the same period for comparison

Buy%20amp%20Hold%20return_zpssep7jmvh.pn

Buy & Hold is better than my strategy over the same period in terms of returns however Buy and Hold is clearly subject to violent drawdowns whereas my strategy attenuates the downside whilst providing a steady upside, and my strategy is active in the market with a live position for only 0.27% of the time that the Buy & Hold strategy is (Minimising event risk).

Although it is clear from the chart that during the last year my strategy has been going sideways.

Link to comment
Share on other sites

  • 1 month later...

My plan was to have started trading at the start of December, I'm torn between running a long only strategy or a long/short strategy. Long/short is considerably more complex to program but it considerably more attractive from a market reality point of view.

 

The principle goal of my work is to identify consistent relationships between a group of markets that identifies a set of conditions that if present means the market is likely to be either higher, or lower. I then program a strategy that defines that relationship and the action I take as a result. I then go through a process of analysis to assess whether it is a viable trading proposition or not.

 

My long only strategy;

 

I consider closing prices over a number of days for a number of markets. If they are in line with what I consider to be favourable conditions then I enter the market. My current favoured strategy has the following metrics between January 2010 and July 2015;

 

Equity curve

Backtest_zpskmnbspnl.png

 

This equates to 200% over 5.5 years but please do keep in mind that it is not particularly difficult to achieve amazing results in the market when conducting analysis on a look back basis as opposed to look forward.

 

My analysis has involved considering the landscape of net profit changes as profit and stop targets change;

3d_zpsw63fjtsb.png

Considering the metrics of trading a 10 contract strategy between January 2010 and July 2015

10%20contracts_zpsthaxc6ha.png]Perf_zps8f1in3ow.png

Link to comment
Share on other sites

  • 1 month later...

Just a quick update, still working through the strategy testing and optimization phase and currently have a strategy that employs long and short trades with the following metrics for the period 1st January 2010 to 31st July 2015, trading 10 contracts per trade using E-Mini Dow Jones Futures;

 

 

Net%20profit_zpsihframwu.png

 

Ratios_zpsovit7i2l.png

 

Equity%20curve_zpsfqrdua7g.png

 

Trade%20analysis_zpsvk2ycoey.png

 

Annual%20period%20analysis_zpsxvkhffbm.p

 

I need to work through the walk forward testing phase. Much of this work is following the guidance of an excellent book by Robert Pardo;

 

Padro_zps0nqksea2.png

Link to comment
Share on other sites

  • 2 months later...

I can scarely believe another nearly 3 months has passed. I hit a roadblock with coding my algorithm but have cleared that hurdle and now am moving through the testing/walk forward phase. I've just launched the autotrade system to trade live on sim account and while I monitor that I will go through the remaining steps of the testing phase.

 

The strategy has an attractive set of metrics when viewed through the lens of backtesting over the last 9 years, trading 10 contracts at a clip on E-Mini Dow Jones futures using starting capital of $100,000.

 

E2_zpsqjkenmoq.png

 

E3_zpspbczcew9.png

 

E4_zpsvhhkxjed.png

 

E5_zpsrtczv5kn.png

 

 

 

I'm constantly playing devil's advocate with my own work somewhat which has pushed things along nicely.

 

Design a strategy. Test the function of the code. Then test the theory through rigorous testing. If it makes it out the other end then the probability is reasonably high that you have something of value that is robust enough to trade with.

 

One of the aspects I look at is what performance is like during very difficult periods. One such example would be the period 11th October 2007 when the market peaked before very significant declines to 6th March 2009 as a result of a major deleveraging due to the credit crisis. The market (Dow Jones futures) peaked at 14252 on 11th October 2007 and declined to a low of 6509 on 6th March 2009. A drop of 55%. During the same period this strategy returned 6%.

 

BP_zpsdeqkskhp.png

 

If I can establish a strong track record trading this strategy using my own funds then at some point in the future I can consider approaching a proprietary trading firm to see if I could trade their funds on a profit split basis.

Link to comment
Share on other sites

  • 1 month later...

I started trading my automated strategies yesterday. I've spent a lot of time on strategy development, optimization and walk forward testing and now is the time to trade.

 

How do the strategies operate?

 

There are two baskets of strategies rather than one single stategy. One basket for the long side and one for the short side.

 

Using genetic algorithms, I carry out an assessment of how various instruments (futures) are positioned and how they interact over previous hours/days. This process creates a basket of strategies, for each of the long and the short side.

 

Once I have each basket, I conduct walk forward testing to stress test each basket, using a Pessimistic Return On Margin test.

 

For each of the long basket and short basket, I select the best performers on the PROM test, and use these variables for a long strategy and a short strategy, these are automated strategies I designed that run in parallel and complement each other.

 

Each of the long and short strategies assess market conditions, and if these are in line with entry conditions a trade is entered, along with a stop and profit order. Trades are exited, if conditions diverge from entry conditions, or if a set time elapses, or the stop order is filled, or if the profit order is filled.

 

Ill post the 10 year performance of the current long & short strategies being used, later.

 

The market I'm trading is E-mini Dow Futures

Link to comment
Share on other sites

  • 2 weeks later...

10 year historic performance of current strategy (long and short strategies combined) - 1 contract

 

EC_zpszqqvobvt.png

 

So far having started auto-trading my strategies on 3rd June there have not been any trades at all. This seemed odd until I checked previous history showing that in fact there have been some previous periods of a few weeks where there are no trades, not many but they are there, probably hard to see but there are flat line points here, although they look small on this chart that covers 10 years (long trades only).

Yield_zpsj6gwhsad.png

 

 

Whilst these strategies are left to run each day, I'm still researching others.

Link to comment
Share on other sites

That is pretty impressive.

 

Can you say more about what you are doing, and how many trades there were over 10 years?

 

How would the results change if you used 15 years? 20 years?

Link to comment
Share on other sites

What I'm doing is using genetic algorithms to search data to find the most opportune market circumstances under which to trade. Market circumstances takes into account aspects such as, the time of day, recent performance of the traded instrument, recent performance of other instruments (stock index futures, bonds, currencies).

 

For the two strategies represented (one long one short) combined on the performance chart, there were 651 trades during last 10 years (520 longs, 131 shorts). 374 winners 277 losers. 57% winners overall (56% for longs, 60% for shorts). All trades are intraday therefore opened and closed during same day. Trade exit criteria is either of profit target or stop level reached (fixed % of index value), or set time limit reached.

 

I only have 10 years of data, I'll see if I can get 15/20 years from another provider.

Link to comment
Share on other sites

First trade of this new approach on Friday (automated Strategy CXIX YM ET L C C)

 

E-Mini Dow September

20160617%20Trade%201_zpstgrd1al1.png

 

Whilst the trade itself was not a good one, a long entered at 17632 and stopped out at 17508 for a loss of 124 points per contract, the automated strategy did as it should, entering the market when the condition was fulfilled, placing the bracket orders, and exiting as specified by the stop.

Link to comment
Share on other sites

I have an interesting Strategy that identifies and exploits a disorderly deleveraging scenario;

 

A disorderly deleveraging is where growth falls below interest rates, the dollar is strong and debt/income ratios rise, this is accompanied by a sharp reduction in financial market asset values.

 

Such a period occurred between July 2008 and February 2009

 

10 year chart of Dow Jones futures (continuous)

January 2008 to April 2009 indicated

Deleveraging%202_zpsbwmq6rov.png

 

"Deleveraging" strategy 10 year performance

January 2008 to April 2009 indicated

Disorderly%20deleveraging%208733%202_zps

 

Interestingly in "normal" times, it goes sideways...

Link to comment
Share on other sites

4th trade (automated Strategy CXIX YM ET L C C) Sim



E-Mini Dow September


20160629%20Trade%204%20S_zpsutngjy4t.png



An entry at 17429 and exit at 17547 for a gain of 118 points per contract



Unfortunately this is a sim trade and not a trade on my live account. Due to the Brexit volatility my clearing firm Wedbush has increased their margin requirements significantly, for many of the trades I initiate the margin has gone up significantly, beyond the value of my account to finance the trades. Some of my trades are now classed as overnight as they are outside the standard day session 08:30 - 15:15 for futures, even though these are always initiated from 08:00 - 08:30 and were previously treated as intraday. They are also treating working orders as each individually requiring separate margin, even though all working orders are always to close an open position.


Link to comment
Share on other sites

5th trade (automated Strategy CXIX YM ET L C C) Sim



E-Mini Dow September


20160630%20Trade%205%20S_zps76ys2pks.png



An entry at 17635 and exit at 17784 for a gain of 149 points per contract. I've spent a year designing, creating, testing and programming these strategies but did not factor in the probability of huge broker margin increases on stock index future products.



My short term goal is to trade my own funds for a year as proof of concept for my automated trading strategies. Then approach a proprietary trading firm to increase the business to a level that produces great profits. I will continue to trade sim with these strategies for the near term but it's incredibly frustrating to find myself in this position. I believe I have done enough testing to be confident enough to trade live.



The only slightly positive aspect is that I am now forced into a period of real-time testing of my strategies and approach, which is a safer test of my work than trading on a live account.



I've come a long long way to this point so I will find a way to do this.


Link to comment
Share on other sites

Archived

This topic is now archived and is closed to further replies.


×
×
  • Create New...