With the weakening of the Fed’s interest rate hikes, the decline of the impact of covid-19, and the reduction of the marginal effect of regional conflicts, the macro environment in 2023 will be better.
We saw a rebound in the financial market in January. And the performance of the crypto-market is better than that of the stock, gold, bond, and foreign exchange.
More discussion about crypto occurred in TG and DC. Although the halving cycle of BTC will not happen until 2024. However, as a risk asset, BTC will be reflected in advance, which can be seen from the merger of ETH last year and the recent policy shift.
Overall, 2023 will have more chances for short-term traders and long-term investors.
Trading Strategy based on AI Algorithm
Trading begins with the establishment of human society. The prevalence of subjective trading is after the emergence of the secondary market. However, the human nature of greed and timidity becomes the biggest obstacle to the execution of trading strategies.
Therefore, traders started to execute trading through programming.
In the past ten years, we have seen the rise of quantitative trading. A quantitative trading strategy has more data support and is executed by computer programming, but it cannot analyze some indicators that are difficult to quantify. Is there a combination that not only has the advantages of quantitative trading but also has the ability to learn and judge?
In this case, we are excited to introduce a new algorithmic trading strategy developed by Sypool that adopts Artificial Intelligence (AI).
The core of the AI trading strategy is “Efficiency creates α”. It is based on quantitative trading, which is executed by building models, backtesting data, and establishing strategies. This part is no different from traditional quantitative trading. On top of this, we add machine learning, which aims to improve the efficiency of iterative strategies to determine which trading strategy is the most suitable in the current market.
The machine learning part mainly includes reinforcement learning and supervised learning. Reinforcement learning is used for the iteration and supervised learning is used to prevent over-fitting.
For example, suppose the historical performance of strategy A is X, but the market is always changing, and the rate for a certain period is Y, then Sypool AI will record it in the database and compare it with the past data to find out the variable (combination Z) that makes X become Y. Then AI regress the relationship between variable combination Z and differ(X, Y), backtest and summarize. Finally, AI gets a new strategy B for the current market.
The next time the variable combination Z occurs, the system will execute strategy B instead of strategy A. The iteration frequency is very high, which can even satisfy high-frequency trading. This part is realized through intensive learning of market data. The variable combination Z is increased The variable combination Z is increased or decreased by supervised learning to prevent overfitting and the influence of illogically correlated factors.
The new trading strategy will be used to select trade targets with potential upside and provide entry signals.
A brief intro to Sypool
Sypool is an asset management protocol that manages users’ wealth and provides a platform for experienced trading teams to deploy their trading strategies.
We found that many asset management institutions in traditional industries have a high threshold for investors and are not transparent enough. For instance, they will require a minimum investment and they may only disclose the positions quarterly.
So we built Sypool, a Decentralized Asset Management Protocol, without any requirements for investors, anyone can manage their own wealth through Sypool, even if it is one dollar.
Sypool’s trading team
Sypool’s product development team members are all experienced traders. They had worked in well-known hedge funds and delivered some very great performances in quantitative trading.
Over the years, they have collected various trading strategies from traditional markets and accustomed those strategies to the crypto market. And the strategies have proven themselves in the past few years.
Besides the AI trading strategies, here are other two main features which will be added in 2023:
Multi Chains Integration
We believe high-performance layer ones such as Aptos and Sui, and scalability solutions, layer twos are sustainable issues for the infrastructure of blockchains so we are always open to integrating high-performance chains to provide better financial services to Web3 natives.
Some popular chains with high TPS and low gas fees such as Aptos, Sui， BNB Chain, Arbitrum, etc. are our priority in terms of high-frequency trading strategy cost.
After launching our Dapp on multi-chains, Sypool will also integrate more dexs to support multi-chain trading thereby providing more SAPs with different trading strategies cross-chains.
Due to the Luna and FTX event last year, the awareness and demand for decentralization and transparency have increased significantly. So we decided to code an Investment DAO platform for investment groups.
An investment DAO is a decentralized community-driven fund. Managers can set up a vault and investors deposit their tokens to get a share and have voting power on investment decisions.
Details can read:
Sypool is a synthetic asset management protocol built by the professional quantitative trading team.
This is another innovation that we have migrated finance from off-chain to on-chain. This idea was inspired by both off-chain fund companies and packaged asset derivatives. On the one hand, we tokenize fund shares. On the other hand, this share token actually represents a small part of a multi-token pool, thus providing scarce liquidity for the cryptocurrency market.
Therefore, you can consider it as a fund share, or you can consider it as a mirrored token pool asset, or your own portfolio, as well as other new uses that may appear in the future.