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.
1,116 documents
The document describes the challenge of forecasting tick-level prices when autocorrelation appears at several horizons, from milliseconds to days. It compares self-exciting point processes, autoregressive models, long-memory methods, frequency-domain…
The document organizes less common exchange order instructions into practical categories rather than treating every parameter combination as a distinct order type. It describes intermarket sweep orders, which allow execution at a venue without first checking…
The discussion asks whether arbitrage in cryptocurrency markets changes realised volatility estimates. One answer argues that, in a simple continuous-time model, arbitrage may be represented through the drift, while volatility is captured by a separate…
The document asks about market makers’ capital, leverage, returns, and share of trading volume. Its answers emphasize that market makers vary widely, from large banks to small proprietary firms, so there is no single representative operating profile. One…
The document raises questions about how US equity market structure works under Regulation NMS. It focuses on the requirement to protect the national best bid and offer, the handling of orders when another venue displays a better price, and the relationship…
The document considers whether access to non-displayed order information could help a stock day trader anticipate price moves, such as a sharp opening gap. The response distinguishes the possible value of knowing more about hidden orders from whether a…
The document describes the data available to an institutional equities execution desk: full-depth order book events, the broker’s own submitted quotes, links from child orders to parent orders, and client order characteristics such as average daily volume,…
The document asks where to find a cross-country timeline of market microstructure developments, including fragmentation, automated quoting, high-frequency trading, and decimalization. It gives Reg NMS and MiFID as examples of reforms with implementation…
The document discusses how researchers might obtain high-frequency trade records that identify market participants or trading parties. It explains that such identifiers are generally tightly restricted because even anonymized records can reveal sensitive…
The document examines a proposed approach to forecasting whether a future tick return will be positive, negative, or unchanged. It describes converting outcomes into class labels, fitting separate binary logistic regressions, and comparing predicted…
The document concerns a CME corn futures order book file whose FIX messages contain timestamps that appear out of chronological order. The question assumes the file should be replayed in timestamp order to reconstruct the market as it happened, and lists…
The discussion compares delayed tick feeds with historical intraday snapshots and explains why free, high-resolution market data can be difficult to obtain. The responses say that free options may exist for selected markets or through academic access, but do…
The document asks how to build a fast order book from FAST market-data updates when the exchange has already matched trades. The task is reconstruction and aggregation of exchange messages, rather than implementing a matching engine. It therefore involves…
The document examines the claim that machine trading can drive market sell-offs because algorithms react to releases without human judgment. It distinguishes execution algorithms, which carry out orders chosen by people, from systems that make trading…
The exchange distinguishes specialized exchange order instructions from ordinary market and limit orders, naming intermarket sweep orders, Day ISOs, and Hide not Slide as examples. The reply says these instructions can create microstructure advantages,…
The document surveys categories of systems used by hedge funds for automated trading and explains why platform choice depends on the instrument and workflow. It names examples spanning Bloomberg order and execution tools, futures-focused systems,…
The document describes bid-ask bounce in high-frequency data: trades alternating between bid and ask can make observed prices jump around the underlying market value. Taking log returns of midprices and features may reduce some nonstationarity, but the…
The discussion examines whether an equity market maker can operate without competing on ultra-low latency. It suggests that slower strategies may be more viable in less liquid stocks, where a participant can post quotes and hold inventory for longer periods.…
The document compares pegged orders with a cancel-and-replace approach for trading through a private forex dealer. Its answer frames the choice around how often the strategy updates orders: frequent repricing and cancellation may favor sending market orders,…
The discussion compares managed languages such as Java and C# with native code such as C++ for algorithmic trading. It argues that the relevant choice depends on the system’s latency needs, workload, and sources of delay. Before switching languages, teams…
The document asks how order-processing latency varies across products and trading infrastructure. It defines realized latency as the interval from local order submission until processing, and points out that latency can matter in equities, options, futures,…
The document explains what RMSE measures when comparing volatility estimates with a latent true volatility. It treats volatility as a parameter and defines estimator error through expected squared distance from that unknown value. Realized volatility can be…
The document asks whether estimates of high-frequency trading profitability are reliable and presents several viewpoints on how to assess the industry. One answer recommends examining public financial statements from listed firms such as Virtu Financial and…
The document outlines a queue-reactive approach to estimating order-flow intensity from high-frequency limit-order-book data. First define event classes such as limit-order insertions, cancellations, and trades. For each price level, group observations by…