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.
9,660 documents
The document asks how to move from evaluating a machine-learning model on historical test data to producing forecasts from newly received broker data. The author describes training several model types on price and fundamental inputs, currently using a random…
The document asks for methods to manage asset allocation across discrete transition states, with attention to liquidity, trading costs, and delayed cause-and-effect relationships. It contrasts moving from cash directly into an asset with transitions between…
The document explains why a trader who steps ahead of a large displayed buy order may have value even when a rising market leaves the trader unfilled. It separates the upward-price scenario into three cases: the order fills before prices rise, fills only…
The document distinguishes market-by-order (MBO) data from market-by-price (MBP) data. MBO records individual orders, while MBP aggregates displayed quantities by price. Exchanges maintain order-level books internally, but traders build the representation…
The document compares keeping strategy research and live trading in one environment with translating the strategy into a separate implementation language. It explains that research often analyzes fixed datasets with vectorized calculations, while live…
The document describes a proposed passive rebate trade in which a trader posts a bid on one marketplace, waits for a fill, then posts an offer at the same price elsewhere. If another participant responds to the displayed price, the trader might complete both…
The document considers how to infer whether a trade was initiated by a buyer or seller when streaming data include the bid, ask, and last trade price. The proposed rule compares the trade price with the contemporaneous quote midpoint, classifying prices at…
The response proposes representing chart patterns as sequences of numerical movements and scanning a rolling window of volume-weighted average price candles for matches. It illustrates the representation with examples such as a double top, head and…
The document asks how to distinguish expected from realized average transaction cost for a market sell order using a limit order book. The questioner walks through consuming visible buy-side depth to fill an order and computes a volume-weighted average price…
The document discusses how to convert a foreign-currency trading gain or loss into a home currency, using a USD/JPY trade as its example. It reports a loss measured in yen and explains that conversion into dollars uses the rate applicable to selling dollars…
The document reproduces a university course description, tentative schedule, reference books, and a long research reading list for quantitative trading. The syllabus spans statistical arbitrage, asset-return properties, statistical models, trading…
The document addresses which options to use for research on S&P 500 index options, comparing contracts on the index, its futures, and exchange-traded funds. It offers a concise liquidity assessment based on electronic-market bid-ask spreads measured in…
The document compares two ways to connect an Interactive Brokers tick feed to analytics and signal generation: a single integrated application that keeps data in memory and writes it to disk, or separate processes that persist feed data for independent…
The document describes a market-making experiment in a video-game item marketplace. It reports that some items appeared to have stable supply and demand, while occasional sales below the best bid created negative spreads. A program repeatedly reposted quotes…
The document discusses how to interpret a participation threshold in a square-root market-impact model. It presents impact as proportional to volatility and the square root of order size relative to daily volume, and notes that the cited framework assumes…
The document asks where forward prices can enter a Monte Carlo derivatives-pricing workflow. It describes a system that accepts calibration instruments, a model choice, initial underlying levels, a yield curve, and a forward curve. The follow-up clarifies…
The document considers how high-frequency market making works when short selling is restricted. It explains two ways a participant may still provide ask-side quotes: an official market maker may receive a short-selling waiver, or another participant may…
The document asks how to choose the trade size that maximizes profit when exploiting a price difference between two constant-product market-making pools. Because each pool’s quoted rate changes with the amount traded, a small trade can receive a different…
The document compares approaches to retrieving SEC 10-K information, depending on whether a researcher needs standardized financial data or the original filings. For historical fundamentals such as income statements, balance sheets, and cash flow statements,…
The document considers whether an equities focused algorithmic firm could earn very high annual returns, including a claim of 100% returns from event driven trading and text mining. Its responses say such outcomes may be possible but emphasize that returns…
The document addresses how to prevent an order-tracking system from recording the same execution more than once when a broker does not provide a separate identifier for each lot. The example includes fills with matching quantities and prices, showing why a…
The document describes a way to extend the standard top-of-book microprice to several price levels. At each level, it weights the ask price by bid-side size and the bid price by ask-side size, then divides the combined weighted prices by total depth across…
The document addresses how to obtain intraday prices for an ETF when standard daily price retrieval is insufficient. It identifies Interactive Brokers as a possible data source through its API and says the API itself does not require payment, while noting…
The document explains why exchanges may not provide idempotency for market orders and describes client-side checks to reduce accidental duplicates. It compares strict matching of past requests, order throttles, checks of consecutive or recent duplicate…