Author Topic: Optimizing Strategies and Curve Fitting  (Read 3518 times)

yell0wbuddha

  • Newbie
  • *
  • Posts: 28
  • Karma: +0/-0
Optimizing Strategies and Curve Fitting
« on: March 14, 2015, 02:33:37 pm »
How are u guys avoiding curve fitting when u are optimizing your strategies?

TradeIdeas_DA

  • Administrator
  • Hero Member
  • *****
  • Posts: 729
  • Karma: +9/-2
    • Trade Ideas LLC
Re: Optimizing Strategies and Curve Fitting
« Reply #1 on: March 16, 2015, 12:13:58 am »
Here's my two cents: the possibility of curve fitting a strategy, the act of designing a strategy to match the best possible performance in the past, is best mitigated by rolling forward with the strategy and taking paper trades to determine how good the strategy is. In my mind curve fitting concerns go away if the strategy continues to hold up given the market conditions. No market lasts forever - no strategy will either without adaptation and continual testing. And none of this is worth much if you don't have good trade management discipline e.g., when to get out of a trade before it gets out of hand.
Follow me on Twitter: http://twitter.com/TradeIdeas

LisaAnwood

  • Newbie
  • *
  • Posts: 2
  • Karma: +2/-0
Re: Optimizing Strategies and Curve Fitting
« Reply #2 on: November 26, 2015, 01:40:29 am »
Hello,

Let me try to explain you the whole process from starting  :)...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 stationary 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.

Here are some important tips you should implement to avoid curve-fitting strategies:

Avoid unreliable simulations. Perhaps the most important thing you need to do in order to avoid curve-fitting a strategy is to completely avoid the use and optimization of systems which cannot be simulated accurately. Simulating systems that trade on time frames lower than 30 minutes or systems with very small take profit and stop loss targets (below 10 times the spread) should be absolutely avoided as the results will not be viable and a lot of curve fitting to past data will most likely take place under optimization. Not only will the results be meaningless but exploitation of backtesting interpolation errors and broker dependency will play a primordial role.

Long testing periods. Optimizations should be carried out for long periods of time, ideally 9-11 years of data should be used for the process in order to ensure that a large amount of market conditions become available. If a simple strategy yields profitable results across a ten year period then the probability of curve fitting is greatly reduced as the system has limited degrees of freedom to artificially “fit” all those different market conditions.

Keep your system symmetric. One of the first ideas new traders have when they start analyzing system development and mathematical expectancy results is to have a separate criteria for entering and exiting short and long trades (for example using an indicator cross at 20 for long entries but 15 for shorts). Although it is true that under past data up and down trends might have developed differently in currencies this cannot be guaranteed to continue in the future as these differences rely on interest rate differentials or such similar macro economic variables that inevitably change through economic cycles. Adding separate criteria for longs and shorts automatically increases the strategy’s degrees of freedom and makes it excessively prone to curve-fitted solutions.

TradeIdeas_DA

  • Administrator
  • Hero Member
  • *****
  • Posts: 729
  • Karma: +9/-2
    • Trade Ideas LLC
Re: Optimizing Strategies and Curve Fitting
« Reply #3 on: November 26, 2015, 08:30:32 am »
Well said!

We are working on greatly expanding the time frame of our historical data from the current 90 days to as much as 5 - 10 years. Remember that most systems tapping into historical data do so at the individual stock level vs. the event-based backtesting that Trade Ideas uses - where knowing the individual stocks ahead of simulations is not required.

With Holly, our Machine Intelligent system, the proof is in the pudding. Each day the algorithms selected either sync well with the overall market and their intended targets to produce profits or they do not. Seeing this performance data in real-time is essential for determining which algos are capturing alpha. In the future Holly will be able to reanalyze what takes overnight to do currently in 5 - 10 minutes thereby displaying what algos are best suited for the current day's (current hour's) activity.
Follow me on Twitter: http://twitter.com/TradeIdeas