How does one get a 130% return of investment using quant based neural networks?
This was a discussion found on Linked In:
2% is good enough if you don’t want to make any more and/or you are currently not making 12%. I am always trying to improve my ROI.
I long ago gave up considering 12% as anything but “chimp” quality returns. Remember the 1980s chimp that threw darts at a board to pick stocks. He beat half of the money managers by just producing random results – as I recall.
A professional should be able to “outperform” the long term market (10-12% average). Otherwise, one should try car repair or plumbing.
I consider 30-35% my MINIMUM expected compounded return. A 35% year represents failure for me. By the way, my trading standard is 1 minute liquidity.
id you say you are getting 130% annually compounded consistently in REAL trading (or on paper) over a multi-year period? If so, I’ll give you 30% to manage my money and take the 100%. I’m not greedy. I would be really curious how you are doing it. My current average expected return is only 60%. If I could bump that average return up to 100%, I’d consider taking a long vacation.
Ok, on ROI: higher is better, but how relate on risk? My prefered fitness goal is return / max drawdawn or downside risk .
e are the first rating organization in the Benelux (since 2003) and the european challenger in quality for S&P and Moody’s.
In order to prove the quality of our ratings a public portfolio of the Dutch Stock exchange (AEX) is managed by third parties and supervised by external compliance delivering 130% total yield (compounded 12,52%) since 1.1.2003 whereas the benchmark realised only 5%.or compounded 0,55%.
The concept can be transfered to any index, we already working on a short/long fund based upon the US ratings of US companies in the NYSE
My methods are broken down into maximizing return and minimizing risk subject to my liquidity requirements. Return is expressed in $ (absolute dollar value of investment). Risk is the probability that I won’t get the projected return. Since forecasts are based on a number of underlying assumptions, my sensitivity to any failure in performance as expected in increased when I start to see a decay in any of the underlying assumptions.
My personal liquidity rule of thumb is based on the principle that I want to be able to sell a position in 1 minute or less without substantially affecting the market. If you can’t dump a position IMMEDIATELY then one incurs greater investment risk.
In other words, if my expectations are not supported by reality on the ground then I want to be able to get out without taking a price hit on my selling price. Maximum liquidity REDUCES risk. I know my predictions AREN’T always going to be right. WHEN they are wrong, I want to be able to cut my losses as quickly as possible without incurring a trading penalty.
I guess your 130% is over a 7 year period for an annual compounded ROI of 12.52%. Guess I can’t go on a vacation. However, I am still confused. 12.52% compounded for 7 years is 228%.
What is your ANNUAL compounded ROI? I don’t care how you get it.
I agree with you on maximizing return and minimizing risk and on importance of liquidity. My question was rather: for you guys for each point of ROI how is the max dradown, downside risk or standar dev of returns acceptable ecc…?
For me in developing strategy phase an annual return of 12% with max dradown of 8% is accetable (we use another measure too…) . Then in real time execution with money managment I control this number!
When I talk about ROI, I am talking about NET at end of period – real money. We are not talking hypothetical, but actual results. What accountants call bottom line net (of costs/losses) net (of taxes and everything else).
There is no downside unless you decide to spend some money.
I use neuroshell by ward systems. I don’t implement it’s trading signals religiously, though. I do use it as the basis for my trades and reccommendations athttp://www.cotsignals.com . You can see some examples of how I use it. I combine a neural network trigger with commitment of traders data.
Are you referring to the book by Murray Ruggiero?