Opix Algo analyses the market microstructure by the order flow to obtain data of volume distribution and limit orders to determine the demand/supply-range of market makers. Makers’ order flows is highly informative and have the strongest predictive power for exchange rates and likely reflects key fundamental information. Their order flow has permanent forecasting power, whereas order flows originating from the other groups only predict transitory changes in exchange rates. These data can be transmitted to predict high volume areas to facilitate price discovery and develop reliable yield curves.
We believe that order flow is a valid proxy variable for private information and is often used in our research in the forex, equities, bonds and futures markets. Order flow analysis allows us to read the data below:
A key function of Opix Algo is to reduce the FX market’s opaqueness by taking information from the market data from trading venues and creating an aggregated picture of prevailing market conditions that informs execution decisions. This aggregated view also referred to as the aggregate order book, provide estimates on transaction volumes in near real-time.
The main aim of this analysis is to reduce market impact, decrease slippage and to predict future price movements based on the transaction data (price, size and direction) for real-time decision making of minimize execution cost.
Artificial neural networks-based decisions are developed for bidding strategies, risk management, analysing patterns and prediction of performance. The ANN has the ability to learn and generate its own knowledge from data inputs such as orderbooks, bid-ask prices, order flows which try to predict the assets price changes by giving a solution. In time-series problems, the ANN is required to build a forecasting model from the historical data set to predict future data points. In short, ANN model is sufficient to gain insight into the directional change of the market by classifying the demand and supply patterns.
The dynamic time warping (DTW) algorithm is an efficient method to match patterns inside a trading system. DTW is a time-series alignment algorithm for measuring two sequences of vector values by warping the distance until an optimal match between the sequences is found. Opix Algo implements DTW to automate matching and determination of the trading position at the predicted price range.
Experiment | Dataset | Accuracy Rate |
1 | AUD - USD 2019 (Jan - August) | 70% |
2 | EUR - USD 2019 (Jan - August) | 72% |
The results are reported with different filter values. Opix Algo breaks the results down into different trading performance metrics as follows:
check on market orders before they are sent to the markets. They allow for automatically blocking or cancelling orders as soon as trades occur outside defined price thresholds, surpass a maximum size, or post and excess number of orders automatically.
allow users or providers to adjust execution parameters during an execution, often when market conditions change, or the algorithm behaves in an undesirable or unexpected way. This is important, for example, in instances of particularly low liquidity when market makers could dominate trading volume or stop trading altogether.
involve continued monitoring of intraday market and carrying trades with counterparties when limits are breached. Algorithms identify errors and potential issues, analysing the particular scenarios and improve execution strategies and risk controls.
Subscribe to our Newsletter! Receive OpixTrade News & Updates direct to your inbox!
You cannot copy content of this page