Quantitative Finance Stack Exchange is a question and answer site for finance professionals and academics. It's 100% free, no registration required.

Sign up
Here's how it works:
  1. Anybody can ask a question
  2. Anybody can answer
  3. The best answers are voted up and rise to the top

Imagine that prior to entering the market you know beforehand the profit factor of similar situations.

For example:

trades similar to TRADE 1 have yielded a 1.2 pf 
trades similar to TRADE 2 have yielded a 2.0 pf

What is the best way to manage lot sizes for the two different trades relative to each other?

The obvious thing to do is

TRADE 1 Lots = X*1.2
TRADE 2 Lots = X*2.0

but I am not sure if this is the smartest move.

share|improve this question
You need to also consider (1) the variation in returns of each trading scenario, and (2) the covariance among the trading scenarios. – chrisaycock Sep 17 '11 at 4:12
@chrisaycock I had considered this, but I cannot think of a good way to apply those concepts that ends with lot management. – Mike Furlender Sep 17 '11 at 5:13
Did you define a maximum global and per position sharpe ratio and exposition you want to achieve ? – Lliane Sep 17 '11 at 6:13
@Lliane I am not sure how I can do this because my entries/exists are dynamic. I can get an "average profit (or loss)" but not "maximum profit (or loss)" - but say I used that, where do I go from there? – Mike Furlender Sep 17 '11 at 6:22
up vote 5 down vote accepted

Maybe not really an answer, but a justification of your approach. It's likely that your results can be expresses as $$ \mathsf EX_1 = 1.2\text{ and }\mathsf EX_2 = 2 $$ where $X_i$ for $i=1,2$ is a random pf of a situation in a class $i$ (we denote it $S_i$). Your method solves the following problem: given a fixed number of trials we would like to maximize the average pf: $$ \mathsf E(\alpha X_1+(1-\alpha)X_2)\to\max_{\alpha\in[0,1]} $$ which certainly has a solution $\alpha = \frac{1.2}{1.2+2} = 0.375$ without any assumptions on independence or correlation of $X_1$ and $X_2$.

So your answer fits exactly this problem. On the other hand you can also think about the following model: given that you encounter $S_1$ let $p$ be the probability that the $S_2$ will appear before you close your $S_1$ position. So, $p$ is the probability that you will lose $S_2$ if you admit $S_1$.

Once you encounter $S_1$ you should decide if admit it, or wait for the possible better situation $S_2$. Let us admit $S_1$ with a probability $\beta$ - then what is the optimal $\beta$? The possible outcomes are:

  1. you admit $S_1$ (pr = $\beta$), then pf = $X_1$;

  2. you reject $S_1$ (pr = $1-\beta$), $S_2$ appears (pr = $p$), then pf = $X_2$;

  3. you reject $S_1$ (pr = $1-\beta$), $S_2$ does not appear (pr = $1-p$), then pf = $0$;

so expected pf is $$ \mathsf E(\beta X_1+(1-\beta)p X_2)\to\max_{\beta\in [0,1]} $$ so you have in the right-hand side $(1.2-2p)\beta+2p$ that is you always admit $S_1$ if $p<0.6$ and always reject $S_1$ if $p>0.6$.

share|improve this answer

Your Answer


By posting your answer, you agree to the privacy policy and terms of service.

Not the answer you're looking for? Browse other questions tagged or ask your own question.