Subscribe by email
- Automatic Code Generation For Quantopian Platform
- Data-mining And Validating Thousands Of Potential Price…
- Deterministic Machine Design of Trading Systems With Strict…
- Quant Trading Without Neural Networks or Genetic Algorithms
- Fooled by Randomness Through Overfitting And Selection Bias
- The Most Important Performance Measure of Trading Strategies…
- Identifying High Probability Short-term Patterns
- Asset Allocation (23)
- Economic Analysis (82)
- Forex trading (8)
- Market Statistics (323)
- Premium Content (158)
- Premium Signals (9)
- Price Action Lab Alerts (133)
- Price Action Lab Patterns (160)
- Quantitative trading (137)
- Risk Management (49)
- Strategy Synthesis (88)
- Technical Analysis (826)
- Trading Strategies (154)
- Trend following (61)
- Uncategorized (38)
Our proprietary PSI mean-reversion long-only strategy is up 15.5% year-to-date with 5% maximum drawdown and 92.3% win rate. P-value is 0.0369 when the strategy performance is compared to that of random long-only traders.
The bulk of trading strategies are random because they are either overfitted to noise or are the result of selection bias. However, there are ways of even profiting from random strategies. In this article I outline the general idea and … Continue reading
Academics have tried hard for many years to convince traders that they cannot profit from randomness. This has negatively affected markets by lowering liquidity and surrendering exchanges to robots and algos. This biased view is also partly responsible for the … Continue reading
This article combines thoughts from an article by Howard Lindzon and my analysis on the performance of random trading. I argue that monkeys win in this market because the bulk of quant trading is worse than random.
Market timers profit at the expense of buy and hold crowd when there is an orderly crash followed by a V-bottom recovery. The chart below demonstrates these facts.
This latest article in Yahoo! Finance is an example of how data-mining bias still drives Wall Street decisions.
The stocks of Intel and Cisco Systems are still 27% and 54% below their all-time high, respectively, on a dividend adjusted basis. The sluggish recovery in the tech sector for about 14 years after the bubble burst of 2000 is … Continue reading
Quant indicators do not completely eliminate all biases but are on the average more successful than visual charting and other techniques of the classical school of technical analysis. Here is how a quant indicator signaled the surge in Intel stock … Continue reading