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 discussion considers whether broker choice and market access can reduce latency and slippage for a futures trader using frequent round trips. Answers describe services available at the time, including direct market access, configurable order routing,…
The note clarifies two details in a trade impulse signal used in market microstructure analysis. The price term of −0.25 represents a spread in price units, rather than the security’s tick size or the price gap between the best and second-best bids. This…
The discussion considers whether intraday volatility and flash crashes matter to investors who hold portfolios for the long term. It distinguishes ordinary short-lived price moves from effects that can persist through trading costs, market impact, and…
The document asks whether Markowitz portfolio weights can be updated efficiently when each new price observation changes the covariance estimate through a rank-one update. It also asks whether using the previous time step’s weights as the optimizer’s initial…
The document outlines an entropy feature for detecting changes in order imbalance that may relate to adverse selection. It starts with volume bars and measures the share of trades classified as buyer initiated. Historical values are divided into quantile…
The document distinguishes exchange quotes from transaction records by tracing them to the matching engine. Orders, cancellations, and replacements enter the engine; its outputs include trades and order-book updates, from which a provider may derive best bid…
The document describes how to estimate the volume ahead of and behind a limit order when exchange acknowledgements and public market data arrive asynchronously. With an order-by-order feed such as ITCH, a system can identify its own order directly, though it…
This explanation describes how to map market observations into a multilayer perceptron for time-series prediction. Each observation forms an input vector, and every hidden neuron computes a weighted combination of its input features, adds a bias, and applies…
This note considers how much historical data to use when forecasting financial time series, such as predicting the next daily market close. It cautions against choosing a lookback window by searching for the interval whose forecast most closely matches a…
The note explains why a realized-kernel data-cleaning procedure may retain trades and quotes from one exchange, such as the NYSE, rather than combining records from multiple venues. The stated aim is to limit distortions caused by reporting delays:…
The document identifies tick data as the common term for records of individual trades without the identities of the parties. Such records may include timestamps at fine time scales, though the discussion does not compare timestamp precision or describe…
The discussion questions a story in which an ETF price allegedly moved after a hedge fund executive entered a ticker and quantity in a retail brokerage interface, before submitting an order. The accepted answer argues that the anecdote does not establish…
The document separates algorithms that decide what and when to trade from those that carry out orders. Execution algorithms break large instructions into smaller trades to manage market impact and may target benchmarks such as volume-weighted average price…
The document examines the claim that high-frequency trading increases liquidity and argues that the HFT label covers participants with different trading behaviors. A participant who posts a resting limit order supplies executable interest to the order book,…
The discussion sketches a learning path for understanding high-frequency trading systems, from market data and exchange connectivity to algorithms and order management. It names FIX, direct market access, data-feed handlers, networking, and programming as…
The document frames a high-frequency trading problem in Chinese futures: low liquidity and large price fluctuations make order execution difficult, while market data arrives in half-second updates. The author asks how to predict the next tick or improve…
The document compares data-driven prediction with theory-based mathematical modeling and presents a broader three-part taxonomy: statistical, equilibrium-based, and empirical approaches. Statistical methods seek predictive accuracy with limited prior…
The document considers how to update an exponential moving average when a time interval contains no new observation, such as a second with no price update. It presents two choices: leave the EMA unchanged, or carry the previous sample forward and compute…
The document discusses whether orders at the same price should be combined into one displayed quantity or kept as separate records. Its practical recommendation is to retain each active order and its properties, then aggregate quantities by price when an…
The document asks whether a financial transaction tax on equity sales can make high-frequency trading infeasible, using Taiwan as an example where the author associates a stated tax with an apparent lack of HFT. The response argues that the effect depends on…
The document sketches a high-frequency trading system split between real-time execution and analytical work. The real-time side consumes market data, maintains local limit order books, runs strategies, and creates, updates, or cancels orders. The analytical…
The document considers what to trade with an imagined near-zero-latency program. It argues that speed within one data center is less decisive than the speed of communication between venues, so faster code alone does not confer a general trading advantage.…
The document addresses how to open a compressed sample of Deutsche Börse Xetra Level 2 market data on Windows. After decompression, the user has a .dmp file that ordinary programs cannot interpret. The explanation identifies the file contents as a FIX/FAST…
This post explains why a trade record in high-frequency equity data can show a price with more decimal places than the displayed quote tick. The answer describes a market order that executes against multiple resting price levels: part of the quantity trades…