@longandshort If you have time check out this YouTube interview with a hedge fund manager. How to Manage Risk Like a Hedge Fund Manager | Interview with Ryan Worch - YouTube I checked out the fund performance and he doesn't beat every year, but he doesn't lose much. I'm not sure how his benchmark compares with S&P500.
Interesting video. I would say this his process is fairly retail (anyone with some money can call themselves a fund manager, sadly, just like everyone with a brokerage account calls themselves a trader ). To get a good understanding of how sophisticated fund managers actually manage risk, check out this primer (from 2003 but still good): sm20030428.pdf (actuaries.org.uk) Most managers will use a combination of the risk management tools listed below (VaR, factor, and variance/stats are the primary ones). Some snippets from the paper: Review of risk models: these are the main types of risk management techniques used:
I feel generally ok about performance because I have limited my factor and industry exposures -- meaning that I derive almost all of my returns from selection (I am a good stock picker). ytd unlevered (censored acc number: factor risk: risk metrics:
I don't agree with all of your conclusions, but you are trading what you preach and have good results...so carry on and congrats on the good year so far!
Speculator! not trader I refer to myself as a Speculator. Is there not some regulatory body that says whether or not you can manage other peoples money? Or can you just set up a web site and call yourself a hedge fund? We have to define simple. This stuff is too far over my head. The only thing I really picked up on is the explanation of why all the different risk models don't work.
@deaddog It’s just that not all risk management approaches work all the time, so it’s best to look at risk across different dimensions. Probably the best and easiest risk management approach is to use reduce portfolio covariance. You can overlay your stops, but reducing covariance means your individual positions are less correlated to each other. You’d want to compare the covariance of your portfolio vs your benchmark. If everything is correlated just pick the best stock.
With a momentum portfolio don't I want a positive covariance? Isn't my optimal risk control to be out of stocks when the momentum stops? Don't even need simple Pythagoras.
That is why hedge funds take months to get into position. If they bought into any stock all at once, the stock price will skyrocket. By buying at small enough quantities, none of the retail traders have a clue that hedge funds are buying into a stock heavily.
That could be a problem for a momentum strategy. By the time they get their position filled the move is over. Us little guys have the odd advantage.
high covariance means you’re not benefitting from diversification. if you have a high covariance portfolio you should just pick the best stock. Ideally you would only add another position if it was not highly correlated. To some degree they will be correlated but that’s where you compare yourself vs your benchmark.