AlphaNova platform roadmap -- what features do you want?
We'd love to hear your feedback! What features would make the AlphaNova platform more useful for your quantitative research?
Some ideas we're considering:
- Live paper trading mode
- Team competitions
- Integrated notebooks (Jupyter-style)
- Social following/feeds
23 Replies
One thing to watch out for: survivorship bias in the training data. Make sure you include delisted securities.
I think the platform should add a paper trading mode so we can test strategies in a more realistic setting between competitions.
I've found that sector neutrality is a key factor in the scoring. Strategies that are long one sector and short another tend to underperform.
One more thing: the scoring engine uses a held-out test period that you never see. So your validation score is the best you can do.
I ran a quick backtest on this idea and got a Sharpe of about 1.2 before costs. Not bad for a simple strategy.
I think the platform should add a paper trading mode so we can test strategies in a more realistic setting between competitions.
For those new to the platform: start with the tutorial competition. It has a smaller dataset and more forgiving scoring.
Have you considered using PCA to reduce the dimensionality of the feature space? I found that the first 10 components capture 80%+ of the variance.
Market regimes are the elephant in the room. A strategy that works in a trending market will fail in mean-reverting conditions.
Has anyone tried using attention mechanisms for this? The temporal attention weights could tell you which historical periods are most relevant.
Market regimes are the elephant in the room. A strategy that works in a trending market will fail in mean-reverting conditions.
One more thing: the scoring engine uses a held-out test period that you never see. So your validation score is the best you can do.
Have you considered using PCA to reduce the dimensionality of the feature space? I found that the first 10 components capture 80%+ of the variance.
Market regimes are the elephant in the room. A strategy that works in a trending market will fail in mean-reverting conditions.
Good point about overfitting. My rule of thumb: never trust a backtest with fewer than 500 observations in the out-of-sample period.
Turnover control is crucial. My best performing model has a turnover of only 8% daily. High turnover strategies rarely survive transaction costs.
import numpy as np
from scipy import optimize
def max_sharpe_portfolio(returns, rf=0.0):
n = returns.shape[1]
init_w = np.ones(n) / n
bounds = [(0.0, 0.1)] * n
constraints = {'type': 'eq', 'fun': lambda w: np.sum(w) - 1.0}
result = optimize.minimize(
lambda w: -(np.mean(returns @ w) - rf) / np.std(returns @ w),
init_w, bounds=bounds, constraints=constraints
)
return result.x
Be careful with the Kelly criterion for position sizing. Full Kelly is way too aggressive. I use quarter-Kelly in practice.