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 asks which R software implements the multi-scale realized volatility estimator, distinguishing it from the two-scale realized volatility estimator and realized kernel methods available in a cited package. The response says it is not aware of an…
The document considers whether QuickFIX can be used from a C project, with an order management system as the intended application. The answers explain that the canonical QuickFIX implementation is built around C++ features, including message classes,…
The document asks how to extend high-frequency realized volatility from a single day to a multi-day horizon. Its proposed method is to calculate daily realized variance from intraday returns and add the daily variances across the target period. Equivalently,…
The document points to foundational literature for building statistical models of order arrivals and limit order book evolution. It identifies a stochastic order book model, a Markovian model for price dynamics, and a broader survey of high-frequency…
The document challenges the idea that a high-frequency trading desk should simply buy as many processors, memory modules, storage devices, and network links as possible. It outlines an example pipeline for market-data handling, signal generation, and…
The document asks why a limit-order-book model’s function for inventory-related value depends on the terminal number of shares held but not on cash accumulated during trading. The answer links this feature to the exponential utility used in the model and to…
The document explains the phrase “buckshot of IOCs” as a metaphor for sending several immediate-or-cancel orders in an attempt to capture a fleeting trading opportunity. An IOC executes against available liquidity when it reaches the matching engine; any…
The document explains how to interpret realized variance (RV) and two-scale realized volatility (TSRV) calculated from high-frequency log returns. RV is the sum of squared returns over the observation interval; under a continuous-time model without jumps or…
The document surveys competing views on whether high-frequency and algorithmic trading benefit markets. One cited study reports that algorithmic trading improved liquidity as measured by bid-ask spreads. The responses question whether spreads alone capture…
The discussion addresses slow downloads of intraday bar data through the Bloomberg API. The accepted response recommends issuing multiple requests concurrently rather than relying on a single request loop, while assigning each request a distinct identifier.…
The discussion considers how to record trades at high speed while preserving balanced accounting entries. One proposed design uses an append-only log, fast insertion, and periodic summary transactions; for distributed systems, it raises the possibility of…
The document discusses how to estimate realized volatility from irregularly spaced tick data over a rolling window. It distinguishes calendar-time sampling, which uses fixed time intervals, from tick-time sampling, which updates on each transaction or at a…
The document asks how quantitative methods might predict rapid odds changes on betting exchanges shortly before an event, using live back and lay prices and available market depth. Responses frame the setting as a market microstructure problem: order flow…
The document frames a research question about whether movements in an FX pair’s midprice on a higher-volume exchange can predict subsequent movements on a lower-volume exchange. The proposed target is whether the second venue’s midprice rises within a chosen…
The discussion explains why scaling minute-level volatility by the square root of time may not make it directly comparable with volatility estimated from daily observations. It identifies two adjustments: account for volatility that occurs overnight, and…
The discussion argues that strategy longevity has no dependable general half-life. One practitioner describes deploying only a small fraction of many researched models: some were retired after exchanges changed data dissemination or order types, while others…
The document describes a data-quality problem in reconstructing an exchange order book from snapshots and updates. Updates overwrite the size at a price level, and the feed provides both provider-received times and exchange event times. Received times have…
The document points readers seeking to provide liquidity in less active markets toward established market making literature. It names books covering financial market liquidity, algorithmic and high-frequency trading, and practical market microstructure, then…
The document describes a practical way to estimate the order-arrival intensity used in Avellaneda–Stoikov-style market-making models, including parameters such as A and k in the Guéant–Lehalle–Tapia formulation. The proposed procedure samples starting times,…
The document explains why the Zhou estimator for high-frequency intraday volatility can produce negative estimates in finite samples. Its correction term uses adjacent return products, which can be negative, even though the underlying volatility parameter is…
The document discusses how to find a forex broker that supports limit orders, client-to-client matching, and algorithmic trading. Responses distinguish retail brokers that may internalize customer flow from agency or ECN models, where orders can be matched…
The document discusses ways to decode FAST, a protocol used for market data, with an emphasis on speed. One proposed software approach reads exchange XML templates and generates Java parsing code. Generating parsers ahead of time avoids recursive handling of…
The document discusses how long market-making and other limit orders may remain active before cancellation or replacement. It cites U.S. equity quote-lifetime data showing a very broad range, from microseconds to many minutes. Responses emphasize that there…
The document considers whether a model that forecasts market behavior a few seconds ahead makes trading logic straightforward. The questioner describes fitting distributions to level-two order-book data and developing signals for both passive and aggressive…