Kelly Criterion for Trading Size

Document Sample
Kelly Criterion for Trading Size Powered By Docstoc
					I know many people be familiar with Efficient Frontier, but I'm guessing that there are
less investors that know about Kelly Criterion. So what is Kelly Criterion and who's
Kelly? Kelly worked at AT&T, and published his original paper back 1956. Its math is
fairly included in communication and information theory, mostly working with
probabilities. However, behind every one of the maths, there lies an astonishing result:
by placing bet amounts in accordance with Kelly Criterion (originally applied to
horse-race gambling), one can maximize the returns in the long term. This can be a
betting formula which includes been tailored to trading and investing:

K% = ( (b 1) * p - 1) / b = ( b*p - (1-p) ) / b

Win probability (p): The probability that any given trade you create will return a
positive amount.

Win/loss ratio (b) or odds: The entire positive trade amounts divided by the complete
negative trade amounts.

If you think of b since the likelihood of b-to-1, payout of b when betting 1 unit of
greenbacks, the numerator is simply the mean price of expected payout, or perhaps
the so-called "edge". Therefore, K% could be expressed as edge/odd. For obvious
reason, you won't want to bet in a game where the expected payout is or negative.

If Kelly Criterion is indeed great, why is that this is simply not heard or used usually
within the investing world. There are a few reasons that prevent it to be used

The volatility of strictly using Kelly Criterion is quite big. Despite the fact that in the
long term, probabilistically speaking your portfolio could have the most return
possible, the good and the bad are so big to become digested by a lot of people.
Therefore, people talk about using "half Kelly" or 1 / 2 of the bet amount calculated
from Kelly Criterion in make an effort to reduce the portfolio volatility.

To use Kelly Criterion, it needs knowing how good you trade options (with regards to
p & b). Obviously, unless you know just how much your "edge" is, the Kelly betting
amount will likely be faraway from the best amount. Estimating and knowing your
edge is a much harder task than calculating the Kelly betting amount.

Despite the mathematical correctness of Kelly Criterion, it's much harder to get such
in practice. Aren't there something that we are able to avoid such a terrific investing
formula? Indeed, there's. This is what I personally learned after investing stocks for
almost ten years now. The riskier the stock/or entry point is, the less amount that you
ought to put in; the safer the stock/or access point is, the more amount that you need
to devote. This really is the spirit of Kelly Criterion that bet should be proportional for
your edge or your supposed advantage. I've been burned by stupid bets so many times
which i finally learned to carefully size every one of my stock transaction. Actually,
sizing of the transaction is equally important if not more than stocks you select. While
almost all of the investment world discusses what to acquire, a smaller amount
attention is invested in just how much you ought to buy. However for every
transaction, it always includes the following elements: what (stock) to buy/sell, when
to buy/sell, and just how much to buy/sell. For successful investing, the 3 elements
has to be carefully chosen. And Kelly Criterion makes it possible to on deciding a
final element: simply how much.More info of stock trading investing

Author Bio