backtesting trading strategies

classic Classic list List threaded Threaded
3 messages Options
Reply | Threaded
Open this post in threaded view
|

backtesting trading strategies

Patrick Burns-2
An R package is available containing some of the code
that was used in the computations of the working paper
described below.  The functions should be of interest
as examples rather than as something to use out of the
box.

The package can be installed with the command:
install.packages('evalstrat',
repos='http://www.burns-stat.com/pages/Freecode')

On Windows it should work on version 2.1.x, 2.2.x and
2.3.x.  If you are on a version older than that, it's time to
update anyway.

In the working papers section of the Burns Statistics website is:

          Random Portfolios for Evaluating Trading Strategies

Abstract: Random portfolios can provide a statistical test that a
trading strategy performs better than chance.  Each run of the
strategy is compared to a number of matching random runs that
are known to have zero skill.  Importantly, this type of backtest
shows periods of time when the strategy works and when it doesn't.
Live portfolios can be monitored in this way as well.  This allows
informed decisions -- such as changes in leverage -- to be made
in real-time.


Patrick Burns
[hidden email]
+44 (0)20 8525 0696
http://www.burns-stat.com

_______________________________________________
[hidden email] mailing list
https://stat.ethz.ch/mailman/listinfo/r-sig-finance
Reply | Threaded
Open this post in threaded view
|

Re: backtesting trading strategies

Roger J. Bos
Patrick,

Thanks for the contribution.  I was intrigued by the idea, but I thought I
would need POP to run the code?  I thought POP was used to create the random
portfolios.  Is POP not necessary, or does it come with a limited-version of
POP.

Thanks,

Roger


On 2/11/06, Patrick Burns <[hidden email]> wrote:

>
> An R package is available containing some of the code
> that was used in the computations of the working paper
> described below.  The functions should be of interest
> as examples rather than as something to use out of the
> box.
>
> The package can be installed with the command:
> install.packages('evalstrat',
> repos='http://www.burns-stat.com/pages/Freecode')
>
> On Windows it should work on version 2.1.x, 2.2.x and
> 2.3.x.  If you are on a version older than that, it's time to
> update anyway.
>
> In the working papers section of the Burns Statistics website is:
>
>          Random Portfolios for Evaluating Trading Strategies
>
> Abstract: Random portfolios can provide a statistical test that a
> trading strategy performs better than chance.  Each run of the
> strategy is compared to a number of matching random runs that
> are known to have zero skill.  Importantly, this type of backtest
> shows periods of time when the strategy works and when it doesn't.
> Live portfolios can be monitored in this way as well.  This allows
> informed decisions -- such as changes in leverage -- to be made
> in real-time.
>
>
> Patrick Burns
> [hidden email]
> +44 (0)20 8525 0696
> http://www.burns-stat.com
>
> _______________________________________________
> [hidden email] mailing list
> https://stat.ethz.ch/mailman/listinfo/r-sig-finance
>

        [[alternative HTML version deleted]]

_______________________________________________
[hidden email] mailing list
https://stat.ethz.ch/mailman/listinfo/r-sig-finance
Reply | Threaded
Open this post in threaded view
|

Re: backtesting trading strategies

Patrick Burns-2
Yes, POP is required to run the backtesting functions
as written.  That is one of a few reasons that people
generally need to modify the code for their own uses.

However, I'm not averse to giving out demos of POP
for people to play with it.  I'm especially lenient to
anyone doing research on random portfolios (or portfolio
optimization).

Pat

roger bos wrote:

> Patrick,
>  
> Thanks for the contribution.  I was intrigued by the idea, but I
> thought I would need POP to run the code?  I thought POP was used to
> create the random portfolios.  Is POP not necessary, or does it come
> with a limited-version of POP.
>  
> Thanks,
>  
> Roger
>
>  
> On 2/11/06, *Patrick Burns* <[hidden email]
> <mailto:[hidden email]>> wrote:
>
>     An R package is available containing some of the code
>     that was used in the computations of the working paper
>     described below.  The functions should be of interest
>     as examples rather than as something to use out of the
>     box.
>
>     The package can be installed with the command:
>     install.packages('evalstrat',
>     repos=' http://www.burns-stat.com/pages/Freecode'
>     <http://www.burns-stat.com/pages/Freecode%27>)
>
>     On Windows it should work on version 2.1.x, 2.2.x and
>     2.3.x.  If you are on a version older than that, it's time to
>     update anyway.
>
>     In the working papers section of the Burns Statistics website is:
>
>              Random Portfolios for Evaluating Trading Strategies
>
>     Abstract: Random portfolios can provide a statistical test that a
>     trading strategy performs better than chance.  Each run of the
>     strategy is compared to a number of matching random runs that
>     are known to have zero skill.  Importantly, this type of backtest
>     shows periods of time when the strategy works and when it doesn't.
>     Live portfolios can be monitored in this way as well.  This allows
>     informed decisions -- such as changes in leverage -- to be made
>     in real-time.
>
>
>     Patrick Burns
>     [hidden email] <mailto:[hidden email]>
>     +44 (0)20 8525 0696
>     http://www.burns-stat.com
>
>     _______________________________________________
>     [hidden email]
>     <mailto:[hidden email]> mailing list
>     https://stat.ethz.ch/mailman/listinfo/r-sig-finance
>
>

_______________________________________________
[hidden email] mailing list
https://stat.ethz.ch/mailman/listinfo/r-sig-finance