Dr. Anna Becker
Co-Founder, CEO. Leads the Artificial Intelligence and Machine Learning teams at EndoTech
EndoTech Spring 2019 Release
3-11-2019

Dear current and potential customers, the first release in 2019 is finally live!  Read more to learn about new versions and features.

We introduce:

V1.4 – for Gold-VIP customers

V1.5 – for Institutional customers


Several updates deserve your particular attention:

1- Reactive Models

The resolution of data, model and signals, is one of the key elements upon which strategy developers must decide.

With regard to data resolution, in general, the more data the better, at least in theory. In practice, many providers offer low-resolution data of 1 to 24 hours, clean and free of charge. While dropping down to 1 minute or lower (every transaction), and ideally to a delta of level 2 (price book) data, is both pricey, hard to store and sometimes not so clean. Add to this the complexity of multiple exchanges and multiple asset types (crypto, forex, ETF’s, stocks) and you can readily see my point of theory versus practice.

We started with 1 day, went to 8 hours, and now are down to 1 minute. In the next version we will add level 1 data (ask/bid/price) and by the end of the year, we’ll reach ultimate Level 2 (price book) data resolution.

With regard to model resolution – again, the more details imported into the model, the better it works … in theory. In practice, reducing dimensions, by deciding from the outset how to normalize your data, takes care of many issues such as overfitting, abnormalities, etc. So, initially, we were fine with the data we had in crypto, but market changes demanded that the model became more sophisticated.

And finally, signals resolution. When considering signals resolution in an ideal world, we think of signals generation immediately when model conditions for entry and exits are met. The higher the frequency – the better.

This is the wrong approach in thin markets like crypto. From the beginning of the bot’s operation, I was concerned about the overall status of the exchanges (hidden/market making bots), the quality of APIs and order execution, and the quality of OMS providers. And I was right to worry.

Order execution costs =commissions + spreads + slippage. This has been our main concern for the last few months. First, we resolved commissions by moving from Taker to Maker fees. Then we introduced smart limit orders to improve slippage. Despite this, slippage still happened, and the process of ensuring that all orders were executed properly was a long struggle. A thin market is a thin market. And before we decided to increase signal resolution, we had to take care of our order execution infrastructure.

Example of v1.5 vs 1.3. Entering with Buy at 107.00 vs at 114.00.

But the higher the resolution the higher the number of fake signals. At this point, EndoTech’s AI/ML decided to introduce optimized filters to reduce fake signal frequency.

2- Connectivity

In v1.4 and v1.5 we introduce direct connectivity to Binance, Bitfinex, and Kraken. We are connected to numerous crypto exchanges through third-party providers, so clients can submit requests to support@endotech.io to check connectivity is available to specific exchanges and we follow up by developing direct connectivity once we see enough interest in it.

 

3- Forex and CFD integration

We came from Fiat and we are returning to Fiat. Given the recent down-cycle in crypto-volatility and liquidity, we decided to add more markets. We view Endotech as a provider of risk-reduction, systematic trading technology in high-volatility (return) markets. From v1.4, we will be adding markets that display high volatility and leveraged markets.

We are now connecting to two major forex brokers’ API’s. Meanwhile, signals of ΣTBot Forex, that are a mix of 5 & 15 min reversal and breakout systems will be available on our website shortly (in the next mini update).

 

HYPOTHETICAL PERFORMANCE RESULTS HAVE MANY INHERENT LIMITATIONS, SOME OF WHICH ARE DESCRIBED BELOW. NO REPRESENTATION IS BEING MADE THAT ANY ACCOUNT WILL OR IS LIKELY TO ACHIEVE PROFITS OR LOSSES SIMILAR TO THOSE SHOWN; IN FACT, THERE ARE FREQUENTLY SHARP DIFFERENCES BETWEEN HYPOTHETICAL PERFORMANCE RESULTS AND THE ACTUAL RESULTS SUBSEQUENTLY ACHIEVED BY ANY PARTICULAR TRADING PROGRAM. ONE OF THE LIMITATIONS OF HYPOTHETICAL PERFORMANCE RESULTS IS THAT THEY ARE GENERALLY PREPARED WITH THE BENEFIT OF HINDSIGHT. IN ADDITION, HYPOTHETICAL TRADING DOES NOT INVOLVE FINANCIAL RISK, AND NO HYPOTHETICAL TRADING RECORD CAN COMPLETELY ACCOUNT FOR THE IMPACT OF FINANCIAL RISK OF ACTUAL TRADING. FOR EXAMPLE, THE ABILITY TO WITHSTAND LOSSES OR TO ADHERE TO A PARTICULAR TRADING PROGRAM IN SPITE OF TRADING LOSSES ARE MATERIAL POINTS WHICH CAN ALSO ADVERSELY AFFECT ACTUAL TRADING RESULTS. THERE ARE NUMEROUS OTHER FACTORS RELATED TO THE MARKETS IN GENERAL OR TO THE IMPLEMENTATION OF ANY SPECIFIC TRADING PROGRAM WHICH CANNOT BE FULLY ACCOUNTED FOR IN THE PREPARATION OF HYPOTHETICAL PERFORMANCE RESULTS AND ALL WHICH CAN ADVERSELY AFFECT TRADING RESULTS.

Past performance, real or hypothetical, is not indicative of future results.

There is a risk of loss in virtual currency trading.

Leave a Reply

Your email address will not be published. Required fields are marked *