ES Trading Performance Update

Q

QuantWizard

Just posting an update on the trading performance of my ES trading system. Feedback welcome. Some KPIs follows below.

KPI Model Benchmark
===================== ========== ===========

GENERAL:
Initial #contracts: 1 1
Current #contracts: 1 1
Capital invested: 5000.00 5000.00
Current equity: 17956.72 5525.56

ROI:
ROI 1 day (%): -13.73 -0.07
ROI 1 week (%): 4.62 0.36
ROI 1 month (%): 10.72 -0.06
ROI 3 months ($): 47.75 1.98
ROI 6 months (%): 49.74 2.43
ROI 1 year (%): 78.12 5.26
ROI total (%): 259.13 10.51
ROI YTD (%): 44.67 0.97
RORAC: 8.25 5.50
Ann. ret (%): 86.04 4.97

RISK MEASURES:
Ann. volatility (%): 34.03 3.62
Modigliani measure: 0.45 0.19
Alpha: 234.96 0.0
Ann. Sharpe ratio: 1.99 1.36
Pr. Sharpe ratio (95%): 1.98 1.35
Max drawdown (%): 26.05 2.49
Win/Loss ratio: 1.27 N/A
Profit/Loss ratio: 1.47 N/A
Beta: 0.60 1.0
R Squared: 0.13 1.0
VaR (5%, 1 Month): 15.28 1.67
 
The performance of your system is very impressive with so many months of trading. Good for you. I understand that your strategy is your intellectual property and you have to keep it in secret. But would you mind providing us with a brieft introduction of your system? Thanks!
 
I can share "two secrets": Figure out when market sentiment will change (i.e. trend following vs. mean reversal), and also use the same knowledge to decide when NOT to trade - like most other algorithms, there are periods where they don't work very well and where you usually blow your account. In other words, avoiding losses are just as important as making profit. The algo itself is quite sophisticated, but not built on any theory - it's basically a result of trial-and-error and continuous improvement. Textbook theories make so many assumptions, most of which don't hold in reality (e.g. random-walk, EMH etc.), so I merely see these as sources of inspiration rather than a basis for algo development.
 
Thanks!

QuantWizard said:
I can share "two secrets": Figure out when market sentiment will change (i.e. trend following vs. mean reversal), and also use the same knowledge to decide when NOT to trade - like most other algorithms, there are periods where they don't work very well and where you usually blow your account. In other words, avoiding losses are just as important as making profit. The algo itself is quite sophisticated, but not built on any theory - it's basically a result of trial-and-error and continuous improvement. Textbook theories make so many assumptions, most of which don't hold in reality (e.g. random-walk, EMH etc.), so I merely see these as sources of inspiration rather than a basis for algo development.
More...
 
You must log in or register to reply here.
Back
Top