Read on Twitter
Backtesting and optimization
--------------------------------
Lot of talks about optimization recently& hence a small thread
Even though am not a big fan of optimizing a system, there are certain things a trader can do to come out as a marginal winner in optimization
1/13
--------------------------------
Lot of talks about optimization recently& hence a small thread
Even though am not a big fan of optimizing a system, there are certain things a trader can do to come out as a marginal winner in optimization
1/13
1. If one has couple of indicators in the trading system or just a simple PA concept, optimizing can eliminate useless rules and parameters. Good news !!
2. Optimization is also useful to determine whether certain kind of stops are useful
2/13
2. Optimization is also useful to determine whether certain kind of stops are useful
2/13
3. For example, a trader can determine if a protective stop can add or deter a system performance.
Often, the distance of trailing stops is too close to the last price, causing premature exits. These determinations can be analyzed more closely with optimization.
3/13
Often, the distance of trailing stops is too close to the last price, causing premature exits. These determinations can be analyzed more closely with optimization.
3/13
4. But the biggest bane of optimization is that we can custom-make any indicator to match the past data perfectly. Especially, after the advent of software that can do this BT task in couple of minutes
Actually, 100s of software are available to do the job for a trader
4/13
Actually, 100s of software are available to do the job for a trader
4/13
5. Because the future is what we are trying to control, most optimization is useless and dangerous as it gives a false sense of confidence. This is a serious issue.
So, it is up to the traderâs penchant for the optimization and its results.
5/13
So, it is up to the traderâs penchant for the optimization and its results.
5/13
6. In my opinion, one has to stop optimizing at some point as the negatives override the positives of optimization by a huge margin.
If one gets stuck in optimization (mostly in the search of holy grail), it is very difficult to come out of it
6/13
If one gets stuck in optimization (mostly in the search of holy grail), it is very difficult to come out of it
6/13
7. There is actually no end to this optimization(curve-fitting) saga.
It is a morphed form of searching for holy grail. It is like a promiscuous man keeping on looking for the best partner & because of his proclivity to get the best (compared to the current partner), ..
7/13
It is a morphed form of searching for holy grail. It is like a promiscuous man keeping on looking for the best partner & because of his proclivity to get the best (compared to the current partner), ..
7/13
..he will keep looking until he dies â somewhere the partner search has to stop
8. Another subtle issue we could see is that the number of hours people spend in charts. It is extremely good to spend time on charts but that can also pave way to optimization eventually.
8/13
8. Another subtle issue we could see is that the number of hours people spend in charts. It is extremely good to spend time on charts but that can also pave way to optimization eventually.
8/13
9. A lot of screen time can help a discretionary trader to learn about market behavior but wont help a mechanical trader that much. As a matter of fact, we dont need to understand market dynamics that much to make money in trading
9/13
9/13
10. Being a mechanical trader, one can devise new systems to start trading once his/her account grows bigger than what the existing single system can handle.
I know many traders who create new system to deploy more money/diversify (not to replace their existing ones)
10/13
I know many traders who create new system to deploy more money/diversify (not to replace their existing ones)
10/13
11. I want to lay down another point here is that to see if the system works on multiple instruments. It should work reasonably well across basket of instruments (with varying returns/risk).
Please remember low vs high beta instruments while making a conclusion about it.
11/13
Please remember low vs high beta instruments while making a conclusion about it.
11/13
12. For example, if it makes losses in other instruments but works only in NF, then most probably, it has been curve fitted for Nifty.
Long story short - there is no silver bullet solution for this problem. Truly, there isn't
12/13
Long story short - there is no silver bullet solution for this problem. Truly, there isn't
12/13
Once we start making money in our system & realize that we can positively control the outcome by MM/position sizing predominantly, this search will stop(hopefully)
Optimization is not all that bad but it would be detrimental for 99% of the traders
Happy trading all !!
13/13
Optimization is not all that bad but it would be detrimental for 99% of the traders
Happy trading all !!
13/13
