3.9 Quantitative Fund

Quantitative Investment Strategies

MetaFX is an ANN (Artificial Neural Network) system independently developed by the X-TECH team from Silicon Valley in North America. For the first time, AI algorithms are applied to the foreign exchange and stock markets, realizing the application of artificial intelligence in the trading field. On the basis of traditional EA trading, the ANN system The security and profitability of transactions have been significantly improved.

ANN(Artificial Neural Network)

Artificial Neural Network (abbreviation: ANN) is a Abstraction and simulation of several basic features such as (Natural Neural Network). After a practical model appeared in the early 1990s, it quickly became popular, and a software system adapted to artificial intelligence was created by simulating biological neural networks.

Features and Advantages of Artificial Neural Networks

• Can fully approximate arbitrarily complex nonlinear relationships.

• All quantitative or qualitative information is stored in equal potential distribution with each neuron in the network, which has strong robustness and fault tolerance.

• The parallel distributed processing method makes it possible to quickly perform a large number of operations.

• Learnable and Adaptive Unknown and Uncertain Systems.

• Ability to handle both quantitative and qualitative knowledge.

The function of ANN

Dedicate data

Artificial intelligence AI with its powerful data capture ability, unforgettable "Memory" and extremely fast computing speed, through deep learning and other methods to analyze and explore market laws, and automatically generate trading strategies accordingly.

Autonomous analysis

The AI ​​system will look for trading opportunities from both technical and fundamental aspects. Deep learning of historical transaction data to discover rules.

Keywords

The AI ​​system adopts a set of Internet keyword capture technology, which can instantly extract tens of thousands of core keywords from websites in a large number of languages ​​at home and abroad.

Automatic modeling

The artificial intelligence algorithm is used to distinguish the positive and negative attributes of the vocabulary, judge the public psychology, and divide the market sentiment into 5 categories. After synthesizing all kinds of information, the AI ​​system will deduce the future trading model.

MetaFX artificial intelligence model

Trend Game Mathematical Quantitative Model

Using mathematical models to describe market behavior is called quantitative finance. Learn from the ancients In the chess form method of the game of Go, mastering the chess form is the key to victory. from price Judging the trend judgment can get the quantitative change from point to surface, so as to establish the trend. Quantitative trading model.

Symbolic Logic Model & Fuzzy Logic Model

Symbolic logic is a form of reasoning that primarily involves the evaluation of predicates. Fuzzy logic relaxes the constraints on the true and false of binary judgments and allows any given predicate to belong to a set of true and false judgments of varying degrees, which is defined in terms of set membership functions. any indicator, be it quantitative indicators, technical indicators, fundamental indicators or indicators of other nature.

Decision Tree Model

There are two main types of decision trees: classification decision trees and regression decision trees. Classification decision trees are characterized by the fact that the output variable is categorical (eg buy, hold or sell), while regression decision trees are characterized by the fact that the output of the characteristic variable is numerical (eg -2.5%, 0%, +2.5%, etc.). Data nature of decision tree training set The quality determines the type of decision tree generation. Algorithms for generating decision trees include C4.5 and Genetic Algorithms.

Neural Network Model

Neural networks are almost the most popular machine learning model for algorithmists. It is composed of a layer of nodes connected to each other between the input and output variables of the routing network.

The model is the core of the algorithmic trading system. In order to make the algorithmic trading system more intelligent, the system should save all historical data of errors, and select a suitable internal model according to this change.

Execution Model

Responsible for completing transactions determined by the model kernel

Supervised Algorithmic Trading Model

Must use the system's endogenous data as a training set, allowing the model to make self-aware adjustments to the environment

Market Analysis Model

Performs the collection, sorting and obtaining of reasonable data to analyze volume and price information for the daily trading profile of a particular security.

Multivariate Model

Manage market quotes based on a single model, balance expected liquidity balances that are not yet used, and then compound management models for multiple markets.

Operating Model

Adaptively increase or decrease the transaction price to deal with the passive attack system of the multi-market

Benchmark Model

According to the gold standard price basis, the price deviation of the gold futures and the price of the standard in the basket are immediately known.

Pan Transaction Data Monitoring Model

Real-time monitoring of global asset managers' trading positions, which adapt to changes in trading entities.

Parameter Optimization Model

Optimizing for important parameters such as price, liquidity, cost, speed and execution possibility, or some combination of them, avoids trade-intensive areas and minimizes the impact of market volatility on profit and loss.

Offensive and Defensive Model

It can identify the counterparty's trading strategy, and the built-in random number model conceals the self-trading strategy.

Price Prediction Model

The price prediction model predicts the future price trend of the historical transaction price trend of stocks, indices, commodities, futures or other tradable varieties that meet high liquidity, no artificial price changes, no extreme news, and are affected by supply and demand.

Recipe Model

After RenAissance Technologies LLC of the United States improved and expanded the Baum-Welch algorithm model in 1988, created the company's most profitable investment portfolio Medallion Fund, and became the world's best-performing hedge fund, algorithmic trading has become more and more popular. favored by the market.