Money Management

Quote from gbos:

Yes I imply that in this example Kelly=90% so 1/10th Kelly is 9%. The fraction of Kelly you are going to use depends on many factors. First is how tolerant are you to drawdowns. For example there is 34% chance that you are halved before doubled if you are using 1 (full) Kelly.

risk0kz.jpg


Other considerations are how uncertain you are about your system statistics. Uncertainties in system’s expectancy, variance etc. must be taken into account before making a meaningful choice.

Also you need to make considerations for black swans by assuming a small but non zero probability to a very damaging event. This also must make you more conservative in choosing the appropriate Kelly fraction.
More...

Apparently we practically would have to take so many factors into consideration to determine a proper size, and several of these factors you mentioned above have not been included normally with the Standard Kelly formula. Some factors would be even allowed to choose certain subjective values. Every trader might use different number of factors. There can be also different ways and calculations, besides yours, to find out the desired fraction, probably having different results against yours. ... ... ...

If anyone still calls this overall approach a Kelly formula/ value, that's fine with me.
 
Quote from OddTrader:

Apparently we practically would have to take so many factors into consideration to determine a proper size, and several of these factors you mentioned above have not been included normally with the Standard Kelly formula. Some factors would be even allowed to choose certain subjective values. Every trader might use different number of factors. There can be also different ways and calculations, besides yours, to find out the desired fraction, probably having different results against yours. ... ... ...

If anyone still calls this overall approach a Kelly formula/ value, that's fine with me.
More...

The theory behind the Kelly approach is one. Same inputs will result the same output. The question ‘which system is better’ will be answered with the same answer by anyone. The question ‘what is the best possible allocation between a number of correlated systems’ will be answered with the same answer by anyone also.

The difference is in the personal preferences of risk. Different people will follow different fractions because of this reason and not because of different calculations. It’s a tradeoff between how quick you want to achieve your goal and how smooth ride you want to have.
 
This is an upgraded version of the The risk of never being unhappy table.

<img src=http://www.elitetrader.com/vb/attachment.php?s=&postid=931029>
Quote from gbos:

The theory behind the Kelly approach is one. Same inputs will result the same output. The question &#8216;which system is better&#8217; will be answered with the same answer by anyone. The question &#8216;what is the best possible allocation between a number of correlated systems&#8217; will be answered with the same answer by anyone also.

The difference is in the personal preferences of risk. Different people will follow different fractions because of this reason and not because of different calculations. It&#8217;s a tradeoff between how quick you want to achieve your goal and how smooth ride you want to have.
More...
 

Attachments

Quote from gbos:

The theory behind the Kelly approach is one. Same inputs will result the same output. The question ‘which system is better’ will be answered with the same answer by anyone. The question ‘what is the best possible allocation between a number of correlated systems’ will be answered with the same answer by anyone also.
More...

Thanks for your feedback.

There are Two Kelly formulas: Original per Kelly's original paper; Simplified as mentioned many times in this thread. Does these Two formulas produce exactly the same value? Which One you use so far?

One calculation can base on theoretical probability values, another historical values. Which One you use so far? If base on theoretical values, the results will be far from practical. If base on historical values, the results wouldn't be the same each time with new data. Which One you use so far?

... ... ...

Telling every trader simply to use "Kelly" won't be enough.


Quote from gbos:

The difference is in the personal preferences of risk. Different people will follow different fractions because of this reason and not because of different calculations. It’s a tradeoff between how quick you want to achieve your goal and how smooth ride you want to have.
More...

If at least One element with subjective value is allowed in a calculation cycle for a formula/ approach, should we say the whole approach can be used to derive objective results? Using Kelly usually is to expect an optimal value for optimal results, allowing trade-off as personal preference is hardly optimal, imo. ... ... ...

"The theory behind the Kelly approach is one." - I agree. That's probability theory.
 
Actually: "The risk of nEver being unHappy" ...

There's no sure thing in trading as in life, but probabilities suffice in most situations.

You may remember the Stone's:
  • "You can’t always get what you want
    But if you try sometimes you just might find
    You get what you need"
Quote from OddTrader:

Surely you know what you're talking about. Keep undating "The risk of Ever being unhappy"! :D
More...
 
Firstly: the input is subjective, the formula is objective.

Secondly: the Kelly criterion is not a probability formula, as it is not related to any probabilistic distribution; one of its "subjective" inputs is the probability of winning.

The Kelly ratio is not a theoretical approach as "1+1=2" in not a theoretical result either ... :)

Quote from OddTrader:

...
If at least One element with subjective value is allowed in a calculation cycle for a formula/ approach, should we say the whole approach can be used to derive objective results? Using Kelly usually is to expect an optimal value for optimal results, allowing trade-off as personal preference is hardly optimal, imo. ... ... ...

"The theory behind the Kelly approach is one." - I agree. That's probability theory.
More...
 
OddTrader,

I don't get you: do you try to be funny? are you arrogant? are you trying to understand money management but you can't? you understand it but disagree?

Please explain. If you think that if I work a real life example from A to Z will be worthy to you, I'll consider spending my time doing it.

Again: I don't get your posts. It's probably me ...
Quote from OddTrader:

http://www.elitetrader.com/vb/showthread.php?s=&threadid=59325&perpage=6&pagenumber=29
More...
 
You must log in or register to reply here.
Back
Top