i love this thread ,very interesting
i will remember
i will remember
- #87
- Dec 9, 2013 2:28am Dec 9, 2013 2:28am
- Joined Aug 2008 | Status: Persist Until Something Happens | 17,898 Posts
Make everything as simple as possible. (A. Einstein)
Those who say it cannot be done should not interrupt those who are doing it
- #88
- Dec 9, 2013 2:41am Dec 9, 2013 2:41am
- Joined Aug 2011 | Status: Trader | 1,327 Posts
{quote} Nice thread mate, but Einstein will be disappointed.
Sometimes the simplest possible is already quite complex. We have to do with it.
No greed. No fear. Just maths.
- #89
- Dec 9, 2013 2:48am Dec 9, 2013 2:48am
- Joined Aug 2008 | Status: Persist Until Something Happens | 17,898 Posts
{quote} Make everything as simple as possible, but not simpler (A. Einstein) Sometimes the simplest possible is already quite complex. We have to do with it.
Those who say it cannot be done should not interrupt those who are doing it
- #90
- Dec 9, 2013 2:51am Dec 9, 2013 2:51am
- Joined Aug 2011 | Status: Trader | 1,327 Posts
{quote} Yes, I know. I was only teasing
No greed. No fear. Just maths.
- #95
- Dec 29, 2013 8:39pm Dec 29, 2013 8:39pm
......thanks good one....
One Must Use A Noble Rapier To Pierce The Veil Of Mystery
- #96
- Jan 2, 2014 10:20am Jan 2, 2014 10:20am
- Joined Aug 2011 | Status: Trader | 1,327 Posts
No greed. No fear. Just maths.
- #98
- Jan 4, 2014 6:05am Jan 4, 2014 6:05am
- Joined Aug 2011 | Status: Trader | 1,327 Posts
{quote} Did you see the Matlab code working out some of the chapter exercises from the book? http://becs.aalto.fi/~ssarkka/ See Publications / Books for the publisher's website. I was thinking of converting some of the code to R where it is easier for me to use in a trading application from my platform. If anyone would like to work on conversion, or general collaboration regarding the text / code, please PM me. Chapter 13 seems like an interesting summary of possible methods, with pluses and minuses of each discussed.
I suck with R. I'm still unable to find good resources to learn it. The online PDF manuals worths nothing.
I hate R having five different libs to do the very same thing. Why QQplot and qqplot? Why fitdist and fitdistr? Why Arima and arima? Why are they incompatible with each other?
Why none of thoses libs have a single line of comment? At least to tell what type the arguments are expected to be. Especially when it is either a list or a vector.
No greed. No fear. Just maths.
- #99
- Jan 4, 2014 6:44am Jan 4, 2014 6:44am
- Joined Aug 2008 | Status: Persist Until Something Happens | 17,898 Posts
Those who say it cannot be done should not interrupt those who are doing it
Hello PipTrapper, good point
In my experience in FX, I tend to believe that ignorance (I talk for myself) is the worse enemy.
Algorithmic trading whether it is High Frequency or simply an Automated Program, represents a mathematical model that trades based on the trading logic developed to exploit a market inefficiency which proved to exist over an extended period of time on a particular asset or financial market.
When I bought my first version of ForexTester to check if I could survive in the market with a manual strategy I realized that I needed specific rules. And that skipping some setups did not save me from going through draw downs. I also noticed that the longer I traded longer the draw down periods became.
All these points made me believe I could not find a strategy that lasts forever. Market changes and we have to adapt to it.
Discretionary trading could not answer many questions, leaving me with a lot of uncertainty about the future.
Since I started to code in MQL4 and read many books on automated trading and strategy optimization I changed my believe of quick profits in the financial market. We need to learn and study to develop the experience and the comprehension to survive when a strategy fails and when market changes.
I do not think that a discretionary trader will be able to explain why he stopped trading a certain strategy and when the market will be back to the same level.
I believe that a more quantitative rather than qualitative approach is needed in order to answer these questions.
Just my experience and point of view.
Best Regards, Francesco.
In my experience in FX, I tend to believe that ignorance (I talk for myself) is the worse enemy.
Algorithmic trading whether it is High Frequency or simply an Automated Program, represents a mathematical model that trades based on the trading logic developed to exploit a market inefficiency which proved to exist over an extended period of time on a particular asset or financial market.
When I bought my first version of ForexTester to check if I could survive in the market with a manual strategy I realized that I needed specific rules. And that skipping some setups did not save me from going through draw downs. I also noticed that the longer I traded longer the draw down periods became.
All these points made me believe I could not find a strategy that lasts forever. Market changes and we have to adapt to it.
Discretionary trading could not answer many questions, leaving me with a lot of uncertainty about the future.
Since I started to code in MQL4 and read many books on automated trading and strategy optimization I changed my believe of quick profits in the financial market. We need to learn and study to develop the experience and the comprehension to survive when a strategy fails and when market changes.
I do not think that a discretionary trader will be able to explain why he stopped trading a certain strategy and when the market will be back to the same level.
I believe that a more quantitative rather than qualitative approach is needed in order to answer these questions.
Just my experience and point of view.
Best Regards, Francesco.