backtesting?

[CarrieW]

how do you do this? do you just go backwards or do you put in a specific date and go forward in real time from that point??

it doesnt make sense to me and I havent ever tried to do it. If I am going backwards then I already know whats happened. I dont see how this would help lol. I'm trying to figure out if it would be benifical to me to be doing backtesting also or just continue to go forward with the market livedemo trading.

if someone could try to explain this concept to me I would appreciate it.

 

Dislike ads? Remove them and support the forum: Subscribe to Fastlane Insiders.

[MJ DeMarco]

You usually need to find a program that will run the BackTesting.

Backtesting involves testing your strategy to see how it would have worked if you started the strategy XX months in the past.

For example, say you hit your target of making profits 2 months in a row using a specific system you created ... for example, "buy when USD hits this and sell when EURO hits this assuming [enter some technical indicator here]"

So now that you've had two months of profits, you want to "backtest" the system to see how the system would have performed if you executed the system for months.

It will extrapolate your system to real, past market conditions. So if Carrie's *system* was executed beginning in 2007, she would have turned that $250 into $4,000. Or, should would have lost all of the money. Its a great tool to see how your chosen system has stood the test of time.

Backtesting gives you a decent indication how your system will work in the future.

 

[CarrieW]

the platform I use has something called a strategy tester, I am assuming this is the backtester.

the way the strategy I am using works you have to be watching the whole time constantly adjusting the fib retracments with every move 40 pips or larger.(unless your doing long term trading) then looking for certian levels to be hit in order to exit. You dont use trailing stop losses.

Is that something a backtester could do?

I dont think I would have any benefit from it the way I am trading(or hope to be) I am going to be pyrmiding the trades at certian levels to get higher returns. so I dont think I would get a realistic enough result to make it worth my while to do at this point.

it just seems too complicated to enter all that info into a program. if its even possible, tho its said in the material its already been backtested and has pretty charts and graphs and all lol.

I cant even make an excel spreadsheet thing lmao.

what do you think? is it enough to go forward in the demo using it or should I try to do some backtesting like on weekends(when market is closed)?

 

[CarrieW]

if I could easily do it then I will.

is what I described with the ever changing fib retracemnts something that I can do in backtesting?

I assume its possible in some fashion becuase it was said to be backtested already. I just dont know if they created a specific program to do it or if the one I have available on my platform will be able to handle the ever changing parameters of what I am doing.

If its not necessary or not possible Id rather find out now then in a few days after pulling all my hair out trying to figure out how to do it lol.

 

Dislike ads? Remove them and support the forum: Subscribe to Fastlane Insiders.

[werbl]

I recently began backtesting, hoping to gain some insight for better stop-loss placement. I trade stocks, the software I use isn't too fancy, so I can't backtest in the program. What I do is print out 6-month or 3-month charts, then just role play. I'll pick my entry point, then use a blank sheet of paper to cover the right side of the chart as I move along, day by day, pretending that I'm in the trade, seeing what happens. Then I use a straightedge, mark my entry and exit points, and roughly calculate my profit or loss for that particular trade. I'll have three or four of the exact same charts printed out, I'll try a slightly different variations of whatever I'm testing on each of the charts, them compare the results when I'm done. Not a very precise method, but it definitely works for me, and has helped.

 

[1320Trader]

Personally the only time I would backtest is if I were building an ATS (automated trading system). I think if you use any discretion in your trading a backtest may yield false results...I also feel the same about paper trading. However, im sure many people use them both successfully.

 

[bflbob]

I use MetaStock for backtesting.
It isn't cheap, but it is better than losing money doing realtime testing.

They have a "Buy" or a "Lease" program for the software.
The buy costs more upfront, but the subscription is cheaper.
The lease costs less upfront, but has a higher monthly fee.

I used to lease, but after leasing for a couple of years, decided to buy.

 

Dislike ads? Remove them and support the forum: Subscribe to Fastlane Insiders.