I had changed the concept of the site and started redesign of it instead of fixing portfolio creation error. We will construct the index of best strategies and everyone would receive same signals for the entire index, like "buying" it. It means, everything would be taken care of, no headaches creating and constantly changing portfolios. When something falls or added to the index, it falls or adds automatically thus making auto-diversification and keeping uo with changes. Just subscribe and forget. Over time probably we would attach it to some APIs to trade auto without sending any emails which also saves time and efforts launching trading platforms and placing trades. -- Best, CEO, Quantrade Ltd., Tadas Talaikis
Sortino ratio (upside potential) added for every strategy. Sortino ratio, like Sharpe ratio, measures risk adjusted returns, but also unlike Sharpe, Sortino ratio penalizes only those returns who are falling below specified target rate. In our case, we had used 6% risk free returns and target of 0, i.e. penalizing only negative returns, when Sharpe also penalizes positive "risk". Read more about Sortino ratio in Wikipedia article.
Added a few RSI based strategies as benchmark, none appears in systems list, but can be explored via strategy constructor.
We're excited to say, now you can get our data through our API!
4 new strategies prefixed with VX added. They are volatility related and many more may be added in this area, but in the past we had the problem with Quandl data not updating on time (before we can place the trade), so we'll just see what we can do here. As always, public updates will catch up over time, because we still are transitioning to updated framework.
WND1 & WND2 strategies were fixed. One of strategies generated not enough signals. After fix, WND1 with Twitter is one of these in the first place with 64%/ year and Sharpe ratio of 0.43. Public updates will catch up over time.
When doing live trading we have noticed a bug that may show incorrect stats for commissions that were dropped from data sources. They will be automatically dropped on our side too. That accounted for around 100 strategies.
PPO1 strategy dropped and replaced by PPO2 which is actually same strategy, just correctly applied.
Correlations feature dropped as I had found most of the strategies are not correlated at all and risk management is very easy here. This also diminish the need to run those stats on billions of records.
I am aware about a bug in "my portfolio" page which deletes portfolio file after deleting some of strategies. Although this is pretty old, it will be fixed after transition to newer software.