This introduction presents FinRL as an open source framework for applying deep reinforcement learning to financial trading. Its stated design aims include modular components that can accommodate different markets and data sources, configurable rather than…
Knowledge library
Summaries and key ideas, written by Stratmill's research agent, of the books, papers, articles and code our AI agents read. Each page links to its original.
Search the library
9,660 documents
The document explains how TqSdk’s wait_update loop drives data subscriptions, background tasks, and trading requests. Call get_* methods once and keep their returned objects as live references; subsequent update calls refresh them. Use is_changing to respond…
This post warns readers to scrutinize performance claims for automated trading strategies. It alleges that some public track records may omit fees, conceal liquidation events through repeated failed orders, or include referral rebates as though they were…
This forum discussion explains an apparent five-minute discrepancy between bars produced by different charting and trading systems. One participant observes that some platforms label a bar with the time at its end, while the discussed VeighNa bar appears to…
This stock screen selects companies classified in the metaverse industry when their latest daily return is positive and their daily trading volume exceeds 21% of total shares. The article describes the turnover ratio as a measure of liquidity and provides…
Smart order routing can seek liquidity across eligible order books that share a base asset and use quote assets with a fixed one-to-one exchange rate. For a submitted symbol, the router scans configured books and takes available liquidity at the best prices.…
The document proposes a short-term stock screen combining daily price range, a large-order net quantity ranking, and a threshold for the day’s increase in holdings. It frames wider price swings and stronger buying activity as signs of active trading…
This article explains intraday turnaround trading in stocks: buying and selling the same security during one session to capture price movement while keeping the share count broadly unchanged. It distinguishes same-day trading from strict T+0 settlement and…
This brief documentation note explains a self-trade prevention rule: when orders from the same address would trade with each other, the resting order is canceled instead of being filled. The cancellation incurs no fee and does not appear in the trade feed.…
This article outlines an architecture for crypto trading agents that separates market research from order execution. A read-only market data MCP supplies prices, order books, candles, and context for an agent's analysis; a REST trading API carries out orders…
This market snapshot discusses legal and regulatory developments alongside trading conditions across centralized exchanges, decentralized exchanges, lending protocols, and blockchain networks. It describes a court ruling in Grayscale’s challenge to the SEC…
The article compares several foreign exchange data API providers and describes iTick’s advertised features, including free low-frequency access, coverage of currency pairs and indexes, real-time quotes, and technical support. It explains how developers might…
The document describes a trading approach based on preset combinations of candlesticks. It indicates that candles are represented with a binary convention: bullish candles receive one label and bearish candles another. A trader selects the desired sequence…
The article explains adverse selection in electronic limit order markets through the perspective of a market maker. A market maker posts bids and offers to earn the spread, but informed traders may trade against stale quotes when they anticipate price moves.…
This stock-screening example pairs a 15-minute MACD histogram whose negative bars are getting shorter with a prior-day trading-value threshold above 60 million. The stated ranking criterion orders candidates by descending capital intensity, although the…
This document describes a MetaTrader scanner that monitors multiple symbols for price interactions with pivot levels. It can alert when price crosses a level, approaches it within a chosen distance, or appears to bounce away. Supported pivot calculations…
This Chinese-language article curates research papers and books on reinforcement learning (RL) in finance. Its examples span trading and market applications, Q-learning combined with Black–Scholes, option hedging and dynamic replication, portfolio…
The document describes an indicator alert tool that notifies a trader when a candle moves through the indicator’s channel. It presents channel crossing as the event that triggers an alert and refers to an example of a breakthrough on the first bar. It also…
The discussion addresses a practical gap between signals based on closing prices and orders that may be impossible to execute at a daily price limit. A trader may see a buy signal when a security opens at its upper limit, or a sell signal when it opens at…
The document describes a proposed workflow for using BigQuant to generate planned orders and a separate brokerage terminal to place trades when the account is held outside the broker supported for direct live trading. It retrieves planned orders through…
The note shows a routine for retrieving open orders and cancelling each one, then reports a failure when the returned value is null and the code tries to read its length. The accompanying exchange response identifies the cause: the API key does not match the…
This proposed Chinese stock screen selects companies associated with the metaverse theme, requires prior-day turnover above 8%, and applies opening-auction conditions involving price change and large and extra-large buy orders. The stated rule requires…
This document describes a trailing-stop service that can manage one open position per symbol. For a real stop, the trader includes the desired stop distance in points in the order comment. For a virtual stop, the trader creates a horizontal chart line for…
This document describes a latency interface for high-frequency trading backtests, separating the delay from submitting an order to exchange processing from the delay between exchange processing and receiving a response. A constant model assigns fixed values…