Discussions about the testing and simulation of mechanical trading systems using historical data and other methods. Trading Blox Customers should post Trading Blox specific questions in the Customer Support forum.
4 posts • Page 1 of 1
Yes, I have a module dedicated to management fees which is part of a more comprehensive module for equity modulations (funding scehdule, management fees, distribution schedule, taxation, etc). I use high-water equity for hedge-fund style management fee structure, like you mention. I aslo combine both performance-based and fixed fees. The fee can be assessed weekly, monthly, quarterly, or annually. I also have fee modules for fixed fees (e.g. $N/period) and overall percentage fee (e.g. 2% equity). Again, these can be assessed over different periods. If I don't have a fee structure, it's fairly simple to drop in a new module.shakyamuni wrote:Does anyone simulate including fees?
The performance fee, on the other hand, is a little trickier and has to implement a high water mark. In reality, the fees are deducted at year end - but this needs to be reflected in the daily NAVs.
What other kind of fee algorithms would be appropriate? I'm making the assumption taxation is on the net before fees, and that fees work on the net before taxes (i.e. they're mutually reductive). Is the assumption always correct? Also, I'm assuming that taxes operate on total equity (that one can really suck and impact LTTF), but management fees are deducted after closing out the position (ie. closed equity). Is the use of closed equity for assessing performance fees correct?
- Management Fee Properties Example
- 20041010_MgmtFees.jpg (53.05 KiB) Viewed 3761 times
I also account for fees in my testing. I am glad c.f. plans to put this functionality in VT. Though a given fee structure tends to affect similar strategies in predictable ways, it is nice to be able to see the difference fees make, especially when marketing a new trading program. If people see that you have a realistic method for accounting for commissions, slippage, and fees the results are more credible (though still only a rough estimate of likely performance).