- Joined
- 8 March 2007
- Posts
- 2,650
- Reactions
- 3,607
I Dump a Good Aussie Movie here worth watching featuring Technical Analysis
" The Bank" on NFLX !
" The Bank" on NFLX !
Dump it Here
- Thread starter Skate
- Start date
- Joined
- 20 July 2021
- Posts
- 11,293
- Reactions
- 15,664
i was quite content @ 0.7 cents just 2 minutes before trading recommenced 0.6 cents might have left me with a part-filled order ( have done that several times on other stocks )
a quick decision that worked out for me this time , not a three year wait ( like i had for BPT )
by the way did i mentioned i am not a trader ..
- Joined
- 21 November 2022
- Posts
- 389
- Reactions
- 988
Mr Skate is a hotbed of subjective Fugazi . I'll have " what is evidence ? " for $500 thanks Alex
- Joined
- 28 December 2013
- Posts
- 6,361
- Reactions
- 24,246
Understanding Backtesting
@Richard Dale recent post, which hyperlinks to the paper “The Three Types of Backtests”, underscores a crucial step for trading success that can significantly enhance the reliability and robustness of any trading system from the outset. The paper explores three primary backtesting methods, ensuring that rigour is applied when testing a trading idea.
The paper emphasises the importance of avoiding common biases and pitfalls in backtesting, which can lead to unreliable strategies. I’m confident that those members with backtesting experience will contribute their insights to this discussion.
Skate.
- Joined
- 28 December 2013
- Posts
- 6,361
- Reactions
- 24,246
The Importance of Backtesting
Backtesting is essential to confirm that a trading idea is valid using historical data. However, the concept of backtesting can be challenging for beginners to grasp, let alone having the knowledge and experience to conduct it properly. Even when backtesting is performed, there is often little understanding of whether it was done correctly or if the results can be relied upon.
Additionally, once you have a backtest result, what metrics should you rely on, and what results must be achieved to consider the trading system profitable?
Skate.
Backtesting is essential to confirm that a trading idea is valid using historical data. However, the concept of backtesting can be challenging for beginners to grasp, let alone having the knowledge and experience to conduct it properly. Even when backtesting is performed, there is often little understanding of whether it was done correctly or if the results can be relied upon.
Additionally, once you have a backtest result, what metrics should you rely on, and what results must be achieved to consider the trading system profitable?
Skate.
- Joined
- 28 December 2013
- Posts
- 6,361
- Reactions
- 24,246
Why Backtest a Trading Strategy?
A term that is often bandied about is “Positive Expectancy,” which essentially means that your trading idea must have a verified statistical trading edge. Fiddling with system parameters, often disguised as optimisation, allows traders to tweak variables to enhance performance.
I’m sure some would have alternative views on fiddling, optimising, or tweaking a trading idea, and there would be some who would likely call out the practice of using poor terminology to describe such methods.
By understanding and properly implementing backtesting, you can develop more reliable and robust trading strategies. However, always be cautious of the potential pitfalls and ensure your backtesting process is as thorough and unbiased as possible, given the software and procedures you follow.
Skate.
A term that is often bandied about is “Positive Expectancy,” which essentially means that your trading idea must have a verified statistical trading edge. Fiddling with system parameters, often disguised as optimisation, allows traders to tweak variables to enhance performance.
I’m sure some would have alternative views on fiddling, optimising, or tweaking a trading idea, and there would be some who would likely call out the practice of using poor terminology to describe such methods.
By understanding and properly implementing backtesting, you can develop more reliable and robust trading strategies. However, always be cautious of the potential pitfalls and ensure your backtesting process is as thorough and unbiased as possible, given the software and procedures you follow.
Skate.
- Joined
- 28 December 2013
- Posts
- 6,361
- Reactions
- 24,246
The Role of Backtesting a Trading Idea
Backtesting is based on the assumption that if a trading idea performed well in the past, it has a good chance of working in the future. Conversely, if the trading idea did not perform well in the past, the common consensus is that it may not work well in the future, which is a reasonable expectation.
What if the Backtest Results are Poor?
If the performance is poor, you may discard the idea or tweak and re-test it, "it’s up to you". You can backtest the idea over any period you like, some prefer longer backtests for confirmation, while others prefer a shorter backtest period to gauge if the trading idea will work with recent shifts in market sentiment. The challenge then becomes how to measure your trading idea on future unknown data, thereby giving you the confidence to develop and apply the trading idea in live trading.
In Other Words
You can code a strategy and find out how it has performed in the past, but this doesn’t guarantee how the idea will perform in the future. If something has performed poorly in the past, it’s unlikely to perform well going forward. Backtesting is as good an indication as you can get, but the reality is that trading is not an exact science.
Skate.
Backtesting is based on the assumption that if a trading idea performed well in the past, it has a good chance of working in the future. Conversely, if the trading idea did not perform well in the past, the common consensus is that it may not work well in the future, which is a reasonable expectation.
What if the Backtest Results are Poor?
If the performance is poor, you may discard the idea or tweak and re-test it, "it’s up to you". You can backtest the idea over any period you like, some prefer longer backtests for confirmation, while others prefer a shorter backtest period to gauge if the trading idea will work with recent shifts in market sentiment. The challenge then becomes how to measure your trading idea on future unknown data, thereby giving you the confidence to develop and apply the trading idea in live trading.
In Other Words
You can code a strategy and find out how it has performed in the past, but this doesn’t guarantee how the idea will perform in the future. If something has performed poorly in the past, it’s unlikely to perform well going forward. Backtesting is as good an indication as you can get, but the reality is that trading is not an exact science.
Skate.
- Joined
- 28 December 2013
- Posts
- 6,361
- Reactions
- 24,246
The Backtesting Process
There are key considerations when it comes to backtesting, and I’m sure others will have their say on their preferred “Backtesting Process.” Especially when it comes to using historical data. Backtesting data "has" to include historical index constituents and delisted securities, to achieve accurate backtesting results.
All backtest metrics tell a different story
There are numerous metrics to consider, such as holding time, total trades, profit/loss ratio, Sharpe ratio, maximum drawdown, and drawdown percentage, just to name a few. With so many metrics to evaluate, it’s understandable why some traders have trouble interpreting backtest results, let alone comparing different versions of the same trading idea.
Skate.
There are key considerations when it comes to backtesting, and I’m sure others will have their say on their preferred “Backtesting Process.” Especially when it comes to using historical data. Backtesting data "has" to include historical index constituents and delisted securities, to achieve accurate backtesting results.
All backtest metrics tell a different story
There are numerous metrics to consider, such as holding time, total trades, profit/loss ratio, Sharpe ratio, maximum drawdown, and drawdown percentage, just to name a few. With so many metrics to evaluate, it’s understandable why some traders have trouble interpreting backtest results, let alone comparing different versions of the same trading idea.
Skate.
- Joined
- 28 December 2013
- Posts
- 6,361
- Reactions
- 24,246
Out-of-Sample Backtesting
After completing a backtest with reasonable results, it is advisable to keep monitoring the trading idea. One effective way to do this is through “out-of-sample testing.” Out-of-sample testing is simply backtesting on unknown data to predict future performance.
Warning
There are dangers in relying too heavily on backtest results, as “overfitting” can sneak up without you even realising the implications it can have on the results.
Skate.
- Joined
- 28 December 2013
- Posts
- 6,361
- Reactions
- 24,246
If You Present Someone with Good Backtest Results
Be prepared for constructive criticism. The first thing they will do is highlight keywords such as overfitting, data-snooping, and other biases to express their concern that these issues may have led to over-optimistic and misleading results.
Always Be Cautious
When backtest results are overly impressive, it pays to remember that even the very best programmers have difficulty achieving similar results. Just be very mindful of the potential pitfalls and ensure your backtesting process is thorough. Sometimes, good enough is not good enough when better is expected.
It's time to hop off my soapbox
Skate.
Be prepared for constructive criticism. The first thing they will do is highlight keywords such as overfitting, data-snooping, and other biases to express their concern that these issues may have led to over-optimistic and misleading results.
Always Be Cautious
When backtest results are overly impressive, it pays to remember that even the very best programmers have difficulty achieving similar results. Just be very mindful of the potential pitfalls and ensure your backtesting process is thorough. Sometimes, good enough is not good enough when better is expected.
It's time to hop off my soapbox
Skate.
- Joined
- 28 December 2013
- Posts
- 6,361
- Reactions
- 24,246
Backtesting in a Nutshell
Backtesting is a crucial step in evaluating how your trading strategy would have performed in the past using historical data, without risking any real money. Even when executed correctly, backtesting alone may not provide enough confidence to trade the strategy live. To further assess the feasibility of live trading, the strategy should be subjected to paper trading.
What is Paper Trading?
Paper trading involves simulating live trading with “virtual money,” which is never at risk. This practice allows you to test your trading strategy in a real-time market environment. Paper trading helps you practice trade execution, understand your emotional reactions to market movements, and observe how your strategy performs under real-time conditions. Unlike backtesting, paper trading accounts for factors like slippage, price gapping, and order execution, which can impact the actual performance of your strategy.
In Summary
Backtesting provides a historical analysis of your trading strategy, while paper trading offers a real-time simulation of how your strategy would perform in the live market. Both are essential tools for developing and refining your trading approach, but they serve different purposes and offer distinct insights.
Skate.
Backtesting is a crucial step in evaluating how your trading strategy would have performed in the past using historical data, without risking any real money. Even when executed correctly, backtesting alone may not provide enough confidence to trade the strategy live. To further assess the feasibility of live trading, the strategy should be subjected to paper trading.
What is Paper Trading?
Paper trading involves simulating live trading with “virtual money,” which is never at risk. This practice allows you to test your trading strategy in a real-time market environment. Paper trading helps you practice trade execution, understand your emotional reactions to market movements, and observe how your strategy performs under real-time conditions. Unlike backtesting, paper trading accounts for factors like slippage, price gapping, and order execution, which can impact the actual performance of your strategy.
In Summary
Backtesting provides a historical analysis of your trading strategy, while paper trading offers a real-time simulation of how your strategy would perform in the live market. Both are essential tools for developing and refining your trading approach, but they serve different purposes and offer distinct insights.
Skate.
- Joined
- 28 December 2013
- Posts
- 6,361
- Reactions
- 24,246
The Value of Community Feedback
Receiving feedback from fellow members is incredibly rewarding. Our community thrives on shared knowledge and experiences, and member contributions certainly add value to this thread. I’ve been receiving wonderful feedback on my trading ideas, which has been immensely helpful. As noted by @Richard Dale, backtesting is a crucial part of strategy development. To address this, I’ve uploaded a short 7-minute YouTube video that may answer questions about backtesting in TradingView. While TradingView’s inbuilt strategy tester may not be as advanced as some other tools, I’ve found it to be adequate for producing the metrics needed to evaluate a trading idea.
TradingView’s “Strategy Tester”
The YouTube video above explains how TradingView’s in-built “Strategy Tester” provides a comprehensive overview of performance metrics. Using TradingView allows you to backtest and paper trade a strategy with ease. The backtest metrics from TradingView are detailed and comprehensive, providing the insights you need to make informed decisions.
Skate.
Receiving feedback from fellow members is incredibly rewarding. Our community thrives on shared knowledge and experiences, and member contributions certainly add value to this thread. I’ve been receiving wonderful feedback on my trading ideas, which has been immensely helpful. As noted by @Richard Dale, backtesting is a crucial part of strategy development. To address this, I’ve uploaded a short 7-minute YouTube video that may answer questions about backtesting in TradingView. While TradingView’s inbuilt strategy tester may not be as advanced as some other tools, I’ve found it to be adequate for producing the metrics needed to evaluate a trading idea.
TradingView’s “Strategy Tester”
The YouTube video above explains how TradingView’s in-built “Strategy Tester” provides a comprehensive overview of performance metrics. Using TradingView allows you to backtest and paper trade a strategy with ease. The backtest metrics from TradingView are detailed and comprehensive, providing the insights you need to make informed decisions.
Skate.
- Joined
- 28 December 2013
- Posts
- 6,361
- Reactions
- 24,246
Skate.
- Joined
- 8 June 2008
- Posts
- 12,774
- Reactions
- 18,677
I would tend to disagree there.I would be paper trading otherwise.
- Joined
- 28 December 2013
- Posts
- 6,361
- Reactions
- 24,246
- Joined
- 28 December 2013
- Posts
- 6,361
- Reactions
- 24,246
Skate.
Richard Dale
Norgate Data
- Joined
- 22 February 2005
- Posts
- 272
- Reactions
- 203
Simple trading strategy testers are very misleading, and causes novices to have a false sense of security that are doing systematic trading. This appears to be one of them.
It's missing really quite crucial systematic trading/backtesting capabilities.
Survivorship bias fail (no delisted stocks), index constituent history fail (only current index constituents), unable to testing multiple strategies, capital allocation, monte carlo, robustness, optimization, adjustments for spinoffs/demergers, to name but a few issues.
Perhaps it's time for readers to devote time to learning something about systematic trading rather than posting on forums - I've given plenty of resources previously.
- Joined
- 28 December 2013
- Posts
- 6,361
- Reactions
- 24,246
@Richard Dale if I gave you a strategy would you Backtest it and do a forum post with your unbiased results
Skate.
Skate.
Similar threads
- Poll
