Pynk Long-Short Gold (B1)
Performance 8/13/20 - 4/24/24
Investment Approach
About Pynk: Pynk (pynk.io) is a crowdsourced investment platform. With thousands of users in over 100 countries, they generate thousands of predictions per day on gold prices, Bitcoin, and NASDAQ. Pynk helps financial institutions and money managers reduce risk and uncertainty when making investment decisions.
Pynk Directional Gold Price Prediction Model: The model identifies optimal conditions for going long or short in gold, or alternatively, holding cash (with neutral sentiment). The model is created by taking filtered sets of aggregated user predictions from Pynk high-performing "super-users". Models are identified and retrained in a roll forward basis to account for change in market regime. The goal of the strategy is to outperform the (long) return for gold in risk-adjusted returns measured via Sharpe ratio.
Pynk Directional Gold Price Prediction Model: The model identifies optimal conditions for going long or short in gold, or alternatively, holding cash (with neutral sentiment). The model is created by taking filtered sets of aggregated user predictions from Pynk high-performing "super-users". Models are identified and retrained in a roll forward basis to account for change in market regime. The goal of the strategy is to outperform the (long) return for gold in risk-adjusted returns measured via Sharpe ratio.
Performance Versus Benchmark (SPDR Gold Shares Total Return) 8/13/20 to 4/24/24
| Strategy Overall | Bench Overall | Strategy YTD | Bench YTD | |
| Abs. Return | -1.6% | 20.0% | 0.4% | 12.4% |
| Rel. Return | -21.6% | N/A | -20.1% | N/A |
| Beta | 0.31 | N/A | 0.01 | N/A |
| Ann. Volatility | 10.9% | 14.0% | 1.0% | 12.8% |
| Sharpe | 0.01 | 0.41 | 10.71 | 2.99 |
| Drawdown | -16.9% | -21.0% | -0.0% | -3.5% |
| IR | -0.23 | N/A | -15.01 | N/A |
| Tracking Error | 13.8% | N/A | 13.9% | N/A |
Historical Summary
| 1 Yr | 2 Yr | 3 Yr | 5 Yr | 10 Yr | |
| Ann. Return | 0.6% | -5.1% | -2.6% | -0.5% | N/A |
| Ann. Volatility | 0.3% | 5.7% | 10.6% | 10.9% | N/A |
Performance Attribution
Risk/Return versus benchmark
Monthly Return Comparison Trailing 12 Months
| May | Jun | Jul | Aug | Sep | Oct | Nov | Dec | Jan | Feb | Mar | Apr | Total | |
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Strategy | -0.0% | -0.0% | 0.0% | -0.0% | -0.1% | 0.1% | -0.0% | 0.0% | 0.0% | 0.0% | 0.0% | 0.2% | 0.2% |
| Benchmark | -1.3% | -2.2% | 2.3% | -1.3% | -4.8% | 7.4% | -0.4% | 0.0% | 0.0% | 0.0% | 1.6% | 8.8% | 9.7% |
| Rel. Return | 1.3% | 2.2% | -2.3% | 1.3% | 4.7% | -7.3% | 0.4% | 0.0% | 0.0% | 0.0% | -1.6% | -8.6% | -9.5% |
Perspective on Performance
This is a long-short model for gold investment based on features from Pynk's gold price prediction data, which comes from thousands of users in over 100 countries.
The model is retrained daily using all prior data.
The model uses a trailing stop loss based on historical volatility to mitigate draw downs.
The backtest considers transaction costs and slippage. See report for additional details.
This model takes the output of aggregated predictions from "super-users" and converts them into 3 signals (long, short, cash).
Allocations are binary (long, short, or in cash).
A minimum of 50 historical predictions are required per super-user.