Artificial intelligence (AI)-based strategies are increasingly getting used in investment and portfolio management. Their contexts, advantages, and outcomes vary widely. in addition to their ethical implications. Yet despite technology that many expect to rework investment management, AI stays a black box for a lot too many investment professionals.
To bring some clarity to the subject, we focused on a selected AI stock trading model and examined what advantages and risk-related costs it will probably bring. Use of proprietary data from AI of tradersan AI trading model from our colleague Ashok Margam and team, we analyzed his decisions and overall performance from 2019 to 2022.
The traders’ AI has few restrictions available on the market positions taken: it will probably enter each long and short positions and reverse positions at any point of the day. However, it exits the market completely by the closing bell of every day, so its positions should not held overnight.
How has the strategy performed over different time periods, trading patterns and volatility environments? And what can this tell us about how AI is likely to be used more broadly in investment management?
Traders’ AI outperformed its benchmark, the S&P 500, over the three-year evaluation period. While the strategy was neutral when it comes to long vs. short, its beta was statistically zero over the timeframe.
Trader AI Model vs S&P 500 Monthly Stock Curve ($10,000 Investment)
The traders’ AI took advantage of moments of upper skewness to attain these results. While the S&P 500 showed a negative skew or a powerful left tail, the AI ​​model showed the alternative: a right skew or a powerful right tail, meaning that the traders’ AI only had a couple of days where they’d very high returns generated.
| AI model | S&P 500 | |||
| Mean | 0.00111881 | Mean | 0.00064048 | |
| Standard design | 0.005669 | Standard design | 0.01450605 | |
| Kurtosis | 11.1665 | Kurtosis | 13.1015929 | |
| Skewness | 1.59167732 | Skewness | -0.62582387 |
So where was the model most successful? Was it higher to go long or short? On high or low volatility days? Does it select the fitting days to take a seat out the market?
As for the latter query, the trader’s AI actually avoided trading on high return days. The company could also be anticipating events with high risk premiums and selecting to not take a position on the direction the market will move.
Traders’ AI performed higher on a market-adjusted basis once they took short positions. On its short days, it averaged 0.13% while the market lost 0.52%. Therefore, the model was higher at predicting down days than up days. This pattern can also be reflected in bear markets, where traders’ AI outperformed in comparison with bull markets.
| Average return of the AI ​​model | Average return of the S&P 500 | |
| When the model is energetic | 0.1517% | -0.0201% |
| When the model sits outside | 0% | 0.8584% |
| If the model is long | 0.1786% | 0.6615% |
| If the model is brief | 0.1334% | -0.5215% |
| If the model is long and Short notice in at some point |
0.1517% | -0.0201% |
| On high volatility days | 0.1313% | -0.0577% |
| On low volatility days | 0.0916% | 0.1915% |
| In bull markets (annual) | 17.0924% | 46.6875% |
| In bear markets (annual) | 20.5598% | -23.0757% |
| In bull markets | 0.0678% | 0.1853% |
| In bear markets | 0.0816% | -0.0916% |
Finally, the AI ​​model performed higher on high volatility days, outperforming the S&P 500 by 0.19% per day on average, while underperforming on low volatility days.
AI model return percentage in comparison with VIX percentage change
All in all, Traders’ AI results show how a selected AI stock trading model can work. Of course, it hardly serves as an indicator of AI applications in investing on the whole. Nonetheless, the incontrovertible fact that it was higher at predicting negative days than rising days, successful when volatility was high, and avoided full trading before major market-moving events are a critical data point. In fact, they point to the big potential of AI to rework investment management.
If you enjoyed this post, do not forget to subscribe .
Photo credit: ©Getty Images / Svetlozar Hristov
