all complete nonsense, first you have to learn how to trade then u backtest.
without learning how to trade all ur money will go to me in no time
without learning how to trade all ur money will go to me in no time
You are using an out of date browser. It may not display this or other websites correctly.
You should upgrade or use an alternative browser.
You should upgrade or use an alternative browser.
backtest for 3 years, blow up in 3 days,
- Thread starter jacksmith
- Start date
Quote from DT-waw:
Clearly, logic or reason is not your strong side.
If this was true, then all possible trading systems will produce no profits in the long run, which is definitely not supported by the evidence, facts.
If you think backtesting is always a flaw, then you automatically assume that each and every system cannot be profitable in the long run since every system obviously can be backtested.
Probably you are unable to come up with any profitable system or rule-based method - that's why you promote such views, I feel sorry for you...
From having read your posts, your "sorrow for me" has the same effect as a 3 year old trying to teach college physics. It is just amazingly funny and pathetic.
A reasonable person does not slam a religious group repeatedly and expect to be taken seriously (no, I am not Jewish). All it did was expose your mind as an uneducated doofus. And that is on a good day
Quote from TraderZones:
From having read your posts, your "sorrow for me" has the same effect as a 3 year old trying to teach college physics.
you're using well known 'ad hominem' attack (just in case ET readers don't know what it is: http://en.wikipedia.org/wiki/Ad_hominem )
why aren't you discussing the substance with me, the core of the issue, backtesting? do you have any arguments against mine?
Let say you are going to bet in which country will John be in any given day.
You have "back-tested" the habits of John and you know that in the last 9 years he has been out of the country for 10 days.
Your back test shows that
after 10 rainy days if the 11th day is Saturday and
it is in the middle of the month and
it is a leap year
the chance of him being outside of the county is 90%.
You have $50,000.
There have been 10 rainy days, the 11th is Saturday, it is the middle of the month and it is a leap year. How much will you bet that John is not in the country?
You have "back-tested" the habits of John and you know that in the last 9 years he has been out of the country for 10 days.
Your back test shows that
after 10 rainy days if the 11th day is Saturday and
it is in the middle of the month and
it is a leap year
the chance of him being outside of the county is 90%.
You have $50,000.
There have been 10 rainy days, the 11th is Saturday, it is the middle of the month and it is a leap year. How much will you bet that John is not in the country?
Quote from ivanbaj:
Let say you are going to bet in which country will John be in any given day.
You have "back-tested" the habits of John and you know that in the last 9 years he has been out of the country for 10 days.
Your back test shows that
after 10 rainy days if the 11th day is Saturday and
it is in the middle of the month and
it is a leap year
the chance of him being outside of the county is 90%.
You have $50,000.
There have been 10 rainy days, the 11th is Saturday, it is the middle of the month and it is a leap year. How much will you bet that John is not in the country?
If you have no estimates of your edge and your odds you cannot bet, at least theoretically there is no way to calculate it.
%Kelly article
The best you can do is bet small to avoid a streak of consecutive losses. Say 5% of you capital at most.
Quote from WyckoffTrader:
Datamine vs dump your variables into a computer and let it spit out your roadmap. That will never work and you must have some trading edge or some dynamic feature that will adjuts to the changing environment. Plus the knowledge of your systems and what environment it will not perform well in and try to filter or pause during this condition.
It ait easy and you will put in your 5000 to 10,000 hours before you see anything robust and not just random luck.
Very wise advice WyckoffTraderâ¦
We all agree poor money management is a killer, building curve fitted solutions is probable and using bad sample testing methods produce poor performing automated solutions. These are the most common causes of new strategies failing in live trading. But there is another area thatâs almost never mentioned and that is that is poor testing methods. As WyckoffTrader said this can lead to 5000 to 10,000 hours (of testing) before you see anything robust and not just random luck
Many traders simply do not understand how to test with price data. Let us start out with the scientific method we all learned in 8th grade science:
âIn order to draw conclusions about outcomes and their causes, data must come from true experiments. True experiments, or randomized field or controlled trials, test specific predictions and rule out alternative explanations. In an experiment, an investigator assigns subjects randomly to experimental and control groups, varies the apparent cause (the independent variable) and looks at the apparent effect (the dependent variable) while holding all other variables constant. Only true experiments can provide evidence of whether an instructional practice works or not.â
About 2500 hours into testing I tried to adopt this method. It was hell to figure out what it meant at first (I had many failures). But when it did work I could finally begin use trade management and make profits in live trading. But the principle is explicit. Trade management principles will have no affect on a trading strategy that has no positive expectancy continuously in live trading. Another other words you canât beat a dead horse and make him work. No amount of trade management can do that.
Here is an example of using this type of testing. Price has a meta-data attached to it. These items are direction, volatility and volume. By testing my strategies in price data that had similar characteristics (direction, volatility and volume) in the in-sample and out-of-sample I was able to isolate what was non-performing in the strategy in testing. Not only that but I now understood when and where my strategies would and would not perform in live trading.
What you are doing using this type of method is data mining in testing. The traderâs objective is to pull as many pieces of the puzzle together as they can to build a unified whole strategy by holding as many things as constant as possible. It is a very complex job to say the least, but the benefits are well worth the tripâ¦
Quote from DT-waw:
you're using well known 'ad hominem' attack (just in case ET readers don't know what it is: http://en.wikipedia.org/wiki/Ad_hominem )
why aren't you discussing the substance with me, the core of the issue, backtesting? do you have any arguments against mine?
you seemed to have ignored your own attacks against the Jewish.
So you think that when you attack people it is ok, and when others do it, you need to dig up dictionary definitions of why they are wrong?
backtesting is futile