Risk Simulation

[Lyric]

I'm looking for something that will graph randomness. I found one that graphs pure coin flips here:

http://www.agribiz.com/merchdiz/cointoss/cointoss.html

But what I want to see is basically the coin flips with a small edge. I want to be able to enter a 51% or 55% chance of winning each flip and see what the graphs look like over trial samples.

Anyone know of software that will do this?

[jukofyork]

[ QUOTE ]
I'm looking for something that will graph randomness. I found one that graphs pure coin flips here:

http://www.agribiz.com/merchdiz/cointoss/cointoss.html

But what I want to see is basically the coin flips with a small edge. I want to be able to enter a 51% or 55% chance of winning each flip and see what the graphs look like over trial samples.

Anyone know of software that will do this?

[/ QUOTE ]
In Excel:

1) Set cell A1 to 0 (your starting amount).
2) Set cell A2 to this formula: =A1+IF(RAND()<=0.51,1,-1)
3) Drag cell A2 down using the "magic box" (no idea what this is really called, as hardly ever use excel...).
4) Then plot the resulting number series as a graph.

To change the p(win) from 51% just change the 0.51 in the excel formula above and re-drag using the "magic box" thing.

Juk [img]/images/graemlins/smile.gif[/img]

[Lyric]

Thank you sir!

[rvg72]

You can also use my ROI Simulator at:

http://www.rvgsoftware.com/roisimulator.zip
http://www.rvgsoftware.com/roisimulatorsource.zip

You need .NET Framework 1.1

it is meant for SNG's but can do what you are looking for and provide a lot of other info.

Setup 1st place = 53% (or whatever) with a payout of 2 and make the buyin 1. Then just select the number of flips for your sample and the number of times that you want to run this and it will provide frequency of ROI results, largest drop frequencies etc.

Here is a screen shot of this type of simulation:



So looking at this, if you want a Risk of Ruin of less than 50% you could play with $6 (assuming $1 per flip) since the lowest point grid shows that your lowest point is -$5 or less 51.7% of the time. (Cum = Cumulative)

I have provided source and this program has been used by a lot of people and I have also released other software, mainly free and 1 commercial app on 2+2 so you don't need to worry too much about spyware in this case.

rvg

[jukofyork]

[ QUOTE ]
You can also use my ROI Simulator at:

http://www.rvgsoftware.com/roisimulator.zip
http://www.rvgsoftware.com/roisimulatorsource.zip

You need .NET Framework 1.1

it is meant for SNG's but can do what you are looking for and provide a lot of other info.

Setup 1st place = 53% (or whatever) with a payout of 2 and make the buyin 1. Then just select the number of flips for your sample and the number of times that you want to run this and it will provide frequency of ROI results, largest drop frequencies etc.

Here is a screen shot of this type of simulation:



So looking at this, if you want a Risk of Ruin of less than 50% you could play with $6 (assuming $1 per flip) since the lowest point grid shows that your lowest point is -$5 or less 51.7% of the time. (Cum = Cumulative)

I have provided source and this program has been used by a lot of people and I have also released other software, mainly free and 1 commercial app on 2+2 so you don't need to worry too much about spyware in this case.

rvg

[/ QUOTE ]
Hey, just a quick thanks for writing this! [img]/images/graemlins/smile.gif[/img]

I didn't know who wrote it, but I have found it very useful since I started playing SNGs: It's a great motivator, as it encourages you to build up a decent sample before jumping to conclusions and it's also a huge "eye opener" for just how bad of a downswing you can expect to get in the long run!

Thanks for this - Juk [img]/images/graemlins/smile.gif[/img]

[rvg72]

No problem at all - I have enjoyed using many of your great programs so happy to add to the 2+2 library of apps.

rvg

[Lyric]

Is there a graphing component to this software?

[rvg72]

No - it seems like you might want to graph a single instance of the simulation. My program compiles results for x number of instances, 100K in the example above.

rvg