Jump to content

Welcome to the new Traders Laboratory! Please bear with us as we finish the migration over the next few days. If you find any issues, want to leave feedback, get in touch with us, or offer suggestions please post to the Support forum here.

  • Welcome Guests

    Welcome. You are currently viewing the forum as a guest which does not give you access to all the great features at Traders Laboratory such as interacting with members, access to all forums, downloading attachments, and eligibility to win free giveaways. Registration is fast, simple and absolutely free. Create a FREE Traders Laboratory account here.

Do Or Die

Smart Optimization And Avoiding Curve-fitting

Recommended Posts

Hi,

 

Optimization refers to the combinatorial search over a range of system input values on price data defined over a fixed number of bars for the cases that produce the best system net profits or some other selected performance variable. It is unavoidable because markets lack stationarity in any form at all and so adaptability is essential for a trading system. By default it is equivalent to curve fitting because you are searching for ‘best fitting’ input parameters. In practice, you can use optimization results to model a more ‘adaptive’ system.

 

Running an optimization can be the easiest way to look for parameters which the market favors in different phases or time intervals. A simple method is to map the changes in optimum input range with a market characteristic such as volatility or trendiness. This in turns helps to build a layer of adaptability or regime analysis which is essential for any trading system.

 

...optimization is only a problem when you fail to take stationarity into account. If you know you are trading within a stable regime, it makes sense to do your best to capture the rules the market is following over that period. My swing trading methodology might best be described as serial optimization: continually hunting for periods of stationary market behavior and trading optimized models derived from those periods.

 

 

Important Points To Avoid Curve-fitting

 

  • Simulations should be close to actual execution: Perhaps the most important aspect of backtesting/optimization is to avoid simulations which cannot be executed in practice similarly as the simulation. For example, in intraday systems, optimizing a strategy which takes profits which are less than 5 times of bid-ask spread will likely be meaningless. Similarly, for EOD systems using the optimized performance on 2008 to trade/test during 2010 can be unprofitable because there was a dramatic shift in trading regimes over these periods.
     
  • Premise based system design: A large number of input parameters or technical indicators lead way to curve-fitting. Using a premise-focused approach helps you keep the parameters minimum and the strategy simple. While modeling a strategy it is easy to get lost on the combination of entry/exit conditions; however, using this approach will help you to focus on the characterstic of market behavior that you intend to exploit in the first place, and which is essential to the ‘edge’ of the system. It is extremely important to code simple “elegant” strategies in order to avoid added complexity which will result in curve fitted solutions.
     
  • Sufficient sample data for simulation: This is a common adage in statistics; there is hardly anything to elaborate. If the period for which the system is tested is small enough, you may need to test it across a large number of symbols. The (expected) number of trades on which the optimization is performed should also be large enough. I prefer minimum 200 trades (ideally more than 1000) which may be tested on more than 100K bars of data. Small sample data increases the chances of curve fitting to a great extent.
     
  • Keep your system symmetric: One of the first ideas new traders have when they start system development is to have a separate criteria for Entry/Exit and Short/Long trades (for example using an indicator parameter 20 for long entries but 15 for shorts). This is because separate criteria increase the number of conditions (and degrees of freedom). If a system from profitability to losses depending only on complicated combination of rules it is less likely to be robust. Even though on daily time frame bull markets tend to have totally different characteristics than bear markets, it is better to have a long(short) only strategy than depending on complicated combinations. Also testing the same strategy over different time frames gives information about its stability.
     
  • Out-of-sample testing periods: A very common practice in system development is go for rolling backtests or walk forward optimization. Certainly the out-sample test doesn’t have to be profitable (as all strategies have profit and draw down periods) but it must at least hold correlation to the draw down depths and profitable periods seen in the past. As a general rule, if an out-sample test shows a MDD (max draw down) more than twice the previous MDD then the strategy may be curve fitted.
     
  • Sharp’ clusters in profits distribution: It is good to have a look at distribution of profits w.r.t. to the change in input parameters. For example, if input 24 gives profitable results but 23 and 25 are unprofitable then it means that the system is ‘instable’ between the input range 23-25. Such system will be unprofitable in actual trading because it depends on particular optimum criterion. Likewise if the system is profitable with input 40 and its profitability gradually decreases as the range is shifted >50 or <30 than it is much more stable with the parameter 40. From a trader's perspective, finding single sharp peak is useless for trading because that result would be instable (too fragile) and not replicable in real trading.

Share this post


Link to post
Share on other sites

Why does the computer optimized strategy performs badly in actual trading?

 

The price data consists of some deterministic signal or pattern (to be used for trading) plus spurious random price movements which are generally referred to as noise.

 

One man's noise is another man's information, so I'll like to elaborate on it. My favorite definition of noise is: what the market does between an entry and exit. For example, you see a pattern formation and buy on breakout at 60 with stop loss at 55 and target 70 respectively. The price action between 55 to 60 is all noise; as well as the price action between 60 to 70. This is a simple example, the noise 'detected' by your system will depend upon the actual strategy inclusive of scaling in/out.

 

The computer optimization of a system’s input parameters on the test data produces performance results that fit the signal AND the noise. The noise process does not repeats in the same way in out-of-sample data or real time (noise is complete randomness!). After an optimization run on the test data, it is difficult to determine whether the input variables chosen from the optimization run have curve fit the noise, which will not be repeated, or have successfully modeled the price signal. The profits on out-of-sample data from choosing the input parameters that generate the best performance results from the optimization fit on the test data will usually disappear in practice, unless the above steps are followed.

Edited by Do Or Die

Share this post


Link to post
Share on other sites

The optimization process I typically follow for stocks (while keeping in mind the points in article):

 

1. Divide the stocks in sets, each having minimum 20 and max 100

- Fast moving stocks like FSLR, NOV, WLP, BIDU etc.

- Slow moving stocks like KO, DOW, ABT, MO etc.

- Low priced stocks between 20 to 40

- High prices stocks above 100

- Large caps

- Stocks which have a thin order book

- Set which I think (before optimization) is best suitable for this particular strategy

- Mixed set containing all of the above

 

2. Run optimization on mixed set. Use a stable parameter from this set to back test (out of sample) on rest of the sets.

 

3. After optimization the strategy should show performance improvement in ALL of the sets. If it does not, I try to find a practical reason related to those stocks like too volatile, more susceptible to gaps ets. If I cannot find a valid reason, keep the original parameters and discard optimization results

 

4. Repeat above steps on each premise used for system development to test whether it adds edge or is just redundant. If the strategy does not makes money on a particular set its ok, but it should not lose money after trading costs.

 

I keep separate sets for intraday and EOD trading systems.

Share this post


Link to post
Share on other sites

Who is the crazy person who is trying to take the topic off thread by asking about copyright? LOL Seriously. This is a topic on system testing and optimisation issues. A topic discussed and written about by hundreds if not thousands of people around the world for more years than I care to remember. There is no copyright on such a topic.

 

This person should be asking themselves why they even broached the irrelevant subject of copyright, and cannot bear to talk about what is really important? We all know, that you know copyright is not relevant here, but it gives an insight into peoples personal hangups and why they fail at trading.

Share this post


Link to post
Share on other sites

Hy,

 

You have a very nice and interesting thread going on here. I am very much interested in optimization but sadly so far I could only find one book related to the topic and to be honest it was a bit above my head. So I would be very thankful to you if you could point me to some material in this subject. Thanks!

Share this post


Link to post
Share on other sites

Try ths book Balazs. Gives a good grounding of all the factors to consider.

 

The Evaluation and Optimization of Trading Strategies by Robert Pardo

 

Trading Sytems That Work by Thomas Stridsman

Share this post


Link to post
Share on other sites
Try ths book Balazs. Gives a good grounding of all the factors to consider.

 

The Evaluation and Optimization of Trading Strategies by Robert Pardo

 

Trading Sytems That Work by Thomas Stridsman

 

 

Thank You very much! I will pick them up as soon as I can.

Share this post


Link to post
Share on other sites
Hi,

 

Optimization refers to the combinatorial search over a range of system input values on price data defined over a fixed number of bars for the cases that produce the best system net profits or some other selected performance variable. It is unavoidable because markets lack stationarity in any form at all and so adaptability is essential for a trading system. By default it is equivalent to curve fitting because you are searching for ‘best fitting’ input parameters. In practice, you can use optimization results to model a more ‘adaptive’ system.

 

Running an optimization can be the easiest way to look for parameters which the market favors in different phases or time intervals. A simple method is to map the changes in optimum input range with a market characteristic such as volatility or trendiness. This in turns helps to build a layer of adaptability or regime analysis which is essential for any trading system.

 

 

 

 

Important Points To Avoid Curve-fitting

 

  • Simulations should be close to actual execution: Perhaps the most important aspect of backtesting/optimization is to avoid simulations which cannot be executed in practice similarly as the simulation. For example, in intraday systems, optimizing a strategy which takes profits which are less than 5 times of bid-ask spread will likely be meaningless. Similarly, for EOD systems using the optimized performance on 2008 to trade/test during 2010 can be unprofitable because there was a dramatic shift in trading regimes over these periods.
     
  • Premise based system design: A large number of input parameters or technical indicators lead way to curve-fitting. Using a premise-focused approach helps you keep the parameters minimum and the strategy simple. While modeling a strategy it is easy to get lost on the combination of entry/exit conditions; however, using this approach will help you to focus on the characterstic of market behavior that you intend to exploit in the first place, and which is essential to the ‘edge’ of the system. It is extremely important to code simple “elegant” strategies in order to avoid added complexity which will result in curve fitted solutions.
     
  • Sufficient sample data for simulation: This is a common adage in statistics; there is hardly anything to elaborate. If the period for which the system is tested is small enough, you may need to test it across a large number of symbols. The (expected) number of trades on which the optimization is performed should also be large enough. I prefer minimum 200 trades (ideally more than 1000) which may be tested on more than 100K bars of data. Small sample data increases the chances of curve fitting to a great extent.
     
  • Keep your system symmetric: One of the first ideas new traders have when they start system development is to have a separate criteria for Entry/Exit and Short/Long trades (for example using an indicator parameter 20 for long entries but 15 for shorts). This is because separate criteria increase the number of conditions (and degrees of freedom). If a system from profitability to losses depending only on complicated combination of rules it is less likely to be robust. Even though on daily time frame bull markets tend to have totally different characteristics than bear markets, it is better to have a long(short) only strategy than depending on complicated combinations. Also testing the same strategy over different time frames gives information about its stability.
     
  • Out-of-sample testing periods: A very common practice in system development is go for rolling backtests or walk forward optimization. Certainly the out-sample test doesn’t have to be profitable (as all strategies have profit and draw down periods) but it must at least hold correlation to the draw down depths and profitable periods seen in the past. As a general rule, if an out-sample test shows a MDD (max draw down) more than twice the previous MDD then the strategy may be curve fitted.
     
  • Sharp’ clusters in profits distribution: It is good to have a look at distribution of profits w.r.t. to the change in input parameters. For example, if input 24 gives profitable results but 23 and 25 are unprofitable then it means that the system is ‘instable’ between the input range 23-25. Such system will be unprofitable in actual trading because it depends on particular optimum criterion. Likewise if the system is profitable with input 40 and its profitability gradually decreases as the range is shifted >50 or <30 than it is much more stable with the parameter 40. From a trader's perspective, finding single sharp peak is useless for trading because that result would be instable (too fragile) and not replicable in real trading.

 

Great, very nice post. Thank you so much

Share this post


Link to post
Share on other sites

Join the conversation

You can post now and register later. If you have an account, sign in now to post with your account.
Note: Your post will require moderator approval before it will be visible.

Guest
Reply to this topic...

×   Pasted as rich text.   Paste as plain text instead

  Only 75 emoji are allowed.

×   Your link has been automatically embedded.   Display as a link instead

×   Your previous content has been restored.   Clear editor

×   You cannot paste images directly. Upload or insert images from URL.


  • Topics

  • Posts

    • Date: 11th July 2025.   Demand For Gold Rises As Trump Announces Tariffs!   Gold prices rose significantly throughout the week as investors took advantage of the 2.50% lower entry level. Investors also return to the safe-haven asset as the US trade policy continues to escalate. As a result, investors are taking a more dovish tone. The ‘risk-off’ appetite is also something which can be seen within the stock market. The NASDAQ on Thursday took a 0.90% dive within only 30 minutes.   Trade Tensions Escalate President Trump has been teasing with new tariffs throughout the week. However, the tariffs were confirmed on Thursday. A 35% tariff on Canadian imports starting August 1st, along with 50% tariffs on copper and goods from Brazil. Some experts are advising that Brazil has been specifically targeted due to its association with the BRICS.   However, the President has not directly associated the tariffs with BRICS yet. According to President Trump, Brazil is targeting US technology companies and carrying out a ‘witch hunt’against former Brazilian President Jair Bolsonaro, a close ally who is currently facing prosecution for allegedly attempting to overturn the 2022 Brazilian election.   Although Brazil is one of the largest and fastest-growing economies in the Americas, it is not the main concern for investors. Investors are more concerned about Tariffs on Canada. The White House said it will impose a 35% tariff on Canadian imports, effective August 1st, raised from the earlier 25% rate. This covers most goods, with exceptions under USMCA and exemptions for Canadian companies producing within the US.   It is also vital for investors to note that Canada is among the US;’s top 3 trading partners. The increase was justified by Trump citing issues like the trade deficit, Canada’s handling of fentanyl trafficking, and perceived unfair trade practices.   The President is also threatening new measures against the EU. These moves caused US and European stock futures to fall nearly 1%, while the Dollar rose and commodity prices saw small gains. However, the main benefactor was Silver and Gold, which are the two best-performing metals of the day.   How Will The Fed Impact Gold? The FOMC indicated that the number of members warming up to the idea of interest rate cuts is increasing. If the Fed takes a dovish tone, the price of Gold may further rise. In the meantime, the President pushing for a 3% rate cut sparked talk of a more dovish Fed nominee next year and raised worries about future inflation.   Meanwhile, jobless claims dropped for the fourth straight week, coming in better than expected and supporting the view that the labour market remains strong after last week’s solid payroll report. Markets still expect two rate cuts this year, but rate futures show most investors see no change at the next Fed meeting. Gold is expected to finish the week mostly flat.       Gold 15-Minute Chart     If the price of Gold increases above $3,337.50, buy signals are likely to materialise again. However, the price is currently retracing, meaning traders are likely to wait for regained momentum before entering further buy trades. According to HSBC, they expect an average price of $3,215 in 2025 (up from $3,015) and $3,125 in 2026, with projections showing a volatile range between $3,100 and $3,600   Key Takeaway Points: Gold Rises on Safe-Haven Demand. Gold gained as investors reacted to rising trade tensions and market volatility. Canada Tariffs Spark Concern. A 35% tariff on Canadian imports drew attention due to Canada’s key trade role. Fed Dovish Shift Supports Gold. Growing expectations of rate cuts and Trump’s push for a 3% cut boosted the gold outlook. Gold Eyes Breakout Above $3,337.5. Price is consolidating; a move above $3,337.50 could trigger new buy signals. Always trade with strict risk management. Your capital is the single most important aspect of your trading business.   Please note that times displayed based on local time zone and are from time of writing this report.   Click HERE to access the full HFM Economic calendar.   Want to learn to trade and analyse the markets? Join our webinars and get analysis and trading ideas combined with better understanding of how markets work. Click HERE to register for FREE!   Click HERE to READ more Market news.   Michalis Efthymiou HFMarkets   Disclaimer: This material is provided as a general marketing communication for information purposes only and does not constitute an independent investment research. Nothing in this communication contains, or should be considered as containing, an investment advice or an investment recommendation or a solicitation for the purpose of buying or selling of any financial instrument. All information provided is gathered from reputable sources and any information containing an indication of past performance is not a guarantee or reliable indicator of future performance. Users acknowledge that any investment in Leveraged Products is characterized by a certain degree of uncertainty and that any investment of this nature involves a high level of risk for which the users are solely responsible and liable. We assume no liability for any loss arising from any investment made based on the information provided in this communication. This communication must not be reproduced or further distributed without our prior written permission.
    • Back in the early 2000s, Netflix mailed DVDs to subscribers.   It wasn’t sexy—but it was smart. No late fees. No driving to Blockbuster.   People subscribed because they were lazy. Investors bought the stock because they realized everyone else is lazy too.   Those who saw the future in that red envelope? They could’ve caught a 10,000%+ move.   Another story…   Back in the mid-2000s, Amazon launched Prime.   It wasn’t flashy—but it was fast.   Free two-day shipping. No minimums. No hassle.   People subscribed because they were impatient. Investors bought the stock because they realized everyone hates waiting.   Those who saw the future in that speedy little yellow button? They could’ve caught another 10,000%+ move.   Finally…   Back in 2011, Bitcoin was trading under $10.   It wasn’t regulated—but it worked.   No bank. No middleman. Just wallet to wallet.   People used it to send money. Investors bought it because they saw the potential.   Those who saw something glimmering in that strange orange coin? They could’ve caught a 100,000%+ move.   The people who made those calls weren’t fortune tellers. They just noticed something simple before others did.   A better way. A quiet shift. A small edge. An asymmetric bet.   The red envelope fixed late fees. The yellow button fixed waiting. The orange coin gave billions a choice.   Of course, these types of gains are rare. And they happen only once in a blue moon. That’s exactly why it’s important to notice when the conditions start to look familiar.   Not after the move. Not once it's on CNBC. But in the quiet build-up— before the surface breaks.   Enter the Blue Button Please read more here: https://altucherconfidential.com/posts/netflix-amazon-bitcoin-blue  Profits from free accurate cryptos signals: https://www.predictmag.com/ 
    • What These Attacks Look Like There are several ways you could get hacked. And the threats compound by the day.   Here’s a quick rundown:   Phishing: Fake emails from your “bank.” Click the link, give your password—game over.   Ransomware: Malware that locks your files and demands crypto. Pay up, or it’s gone.   DDoS: Overwhelm a website with traffic until it crashes. Like 10,000 bots blocking the door. Often used by nations.   Man-in-the-Middle: Hackers intercept your messages on public WiFi and read or change them.   Social Engineering: Hackers pose as IT or drop infected USB drives labeled “Payroll.”   You don’t need to be “important” to be a target.   You just need to be online.   What You Can Do (Without Buying a Bunker) You don’t have to be tech-savvy.   You just need to stop being low-hanging fruit.   Here’s how:   Use a YubiKey (physical passkey device) or Authenticator app – Ditch text message 2FA. SIM swaps are real. Hackers often have people on the inside at telecom companies.   Use a password manager (with Yubikey) – One unique password per account. Stop using your dog’s name.   Update your devices – Those annoying updates patch real security holes. Use them.   Back up your files – If ransomware hits, you don’t want your important documents held hostage.   Avoid public WiFi for sensitive stuff – Or use a VPN.   Think before you click – Emails that feel “urgent” are often fake. Go to the websites manually for confirmation.   Consider Starlink in case the internet goes down – I think it’s time for me to make the leap. Don’t Panic. Prepare. (Then Invest.)   I spent an hour in that basement bar reading about cyberattacks—and watching real-world systems fall apart like dominos.   The internet going down used to be an inconvenience. Now, it’s a warning.   Cyberwar isn’t coming. It’s here.   And the next time your internet goes out, it might not just be your router.   Don’t panic. Prepare.   And maybe keep a backup plan in your back pocket. Like a local basement bar with good bourbon—and working WiFi.   As usual, we’re on the lookout for more opportunities in cybersecurity. Stay tuned.   Author: Chris Campbell (AltucherConfidential) Profits from free accurate cryptos signals: https://www.predictmag.com/   
    • DUMBSHELL:  re the automation of corruption ---  200,000 "Science Papers" in academic journal database PubMed may have been AI-generated with errors, hallucinations and false sourcing 
    • Does any crypto exchanges get banned in your country? How's about other as Bybit, Kraken, MEXC, OKX?
×
×
  • Create New...

Important Information

By using this site, you agree to our Terms of Use.