This is 100% true. Because of the low operating costs, and poor resources a retail trader has, to have a monthly net-winning intraday system is basically printing money. It's rare for a reason. It requires a very abstract and multi-dimensional thought process to hit any pay dirt.
Edward O. Thorp on charting
- Thread starter truetype
- Start date
This is 100% true. Because of the low operating costs, and poor resources a retail trader has, to have a monthly net-winning intraday system is basically printing money. It's rare for a reason. It requires a very abstract and multi-dimensional thought process to hit any pay dirt.
Without revealing anything about your strategy, what do you mean by quantitative trading, and what is the difference between just programming a system based on the classic TA indicators? Do you use machine learning?
You can Google quant trading and get a hundred definitions, so I won't explain it.
No machine learning here. Just simple historical data analysis. A coded system with properly implemented math will outperform one based on TA/indicators.
No machine learning here. Just simple historical data analysis. A coded system with properly implemented math will outperform one based on TA/indicators.
Actually , its not true. Folks need lump sums of money fast for a variety of reasons-- hence they will sell system rather than grinding out long term profits.
There's an entire industry built around a small lump sum in exchange for a large cash flow. Jg wentworth anyone??? LOL!
I know what it is, but there are a hundred definitions as you say, so I wasn't sure what you meant. It could have been a system based on price/volume etc, or company data for stocks.
Quant is nothing more than applying mathematics to data to make decisions in the market.
Ok, but that's also what indicators like RSI, MACD do and which in my opinion are of little use (they just apply a mathematical operation to past data). Quant should probably be defined as something that uses more advanced math and statistics to achieve an edge I think.
It doesn't have to be advanced math. It's all in the way it's implemented. Simple aggregation and fragmentation could provide a good hypothesis, considering it can be executed in real-time
Create your own adaptive techniques
No, you are not quite getting it. Here is an interview I conducted with one of the fathers of quant analysis over a decade ago. It should shed some light.
http://tradingmarkets.com/recent/quants_on_wall_street_the_emanuel_derman_interview-656152.html