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 explains how the semi-variance HAR model (SHAR) extends the heterogeneous autoregressive model of realized volatility to represent asymmetric effects of past returns. It splits the daily realized variance into positive and negative…
The document describes how a trading algorithm might infer another participant’s behavior from changes in displayed order-book liquidity and response timing. In its example, a large bid is partly consumed; a second participant quickly removes the remaining…
The document surveys ways to build a user interface for controlling trading software whose strategy runs in C++. The suggested design is to expose a small service from the strategy host, then let a separate web or desktop client send commands and display…
The document discusses estimating the liquidity parameters in the Avellaneda–Stoikov market-making model from historical limit-order data. The model describes fill intensity at a quote distance from fair value using a baseline intensity A and a distance…
The discussion examines whether rapid algorithmic reactions to public news can overshoot and then reverse, and whether such reversals challenge semi-strong market efficiency. Responses describe plausible mechanisms: automated systems may prioritize…
The document surveys specialized computing used in trading, including FPGA-based ticker plants, network cards designed to reduce software and kernel overhead, and hardware for feed handling and real-time risk. It also describes GPUs for computational work…
The document discusses why some customers continue to route large SPX and VIX option orders through open-outcry pits when electronic trading is also available. The accepted response attributes better execution prices to negotiation between floor traders, who…
The document compares approaches for storing financial time series and serving them to R, with particular attention to tick data. One response argues that database choice matters less than matching storage layout and indexing to query patterns: a clustered…
The document surveys ways to capture, store and later analyze market tick data. It describes collecting Interactive Brokers market updates through an R interface and writing records to files or in-memory time-series objects. For longer-term analysis,…
The document compares annualized standard deviation of returns with realized volatility computed as the square root of summed squared high-frequency returns. It says the realized variance approach can make better use of intraday observations than a standard…
The document examines the assumption that a small order has negligible influence on market observations and price dynamics. That assumption can make historical backtests easier because the recorded order book or other market data are left unchanged by…
The document gives a standard-error approach to estimating how many observations are needed to assess a strategy’s returns. It relates sample size to measured return variability, a chosen confidence level, and the desired width of the confidence interval.…
The document explains a cross-sectional regression used to study whether stock returns in one half-hour interval respond to returns from earlier intervals. For each lag and time interval, the researchers regress the cross-section of current stock returns on…
The document offers guidance for choosing a short applied mathematics project involving financial data and time-series analysis. It recommends aligning the topic with the supervisor’s research interests and adapting an existing empirical paper into a smaller…
The document surveys several approaches to modeling limit order books and market microstructure: agent-based models, point-process models, propagator models, and Markov-chain approaches such as the Queue Reactive model. It describes the strengths associated…
The document points readers to research on estimating and forecasting volatility from very short-interval, tick-by-tick data. It highlights estimators designed to reduce distortion from market microstructure, along with work on realized variance, noise…
The document surveys ways to estimate which kinds of participants are active in a market, while noting that reliable high-frequency identification is difficult, especially in equities. Suggested sources and proxies include weekly futures positioning reports,…
The discussion explains that pegged orders do not necessarily retain first-in-queue priority when prices move. Queue handling depends on the exchange’s order type and matching rules; traders should check venue documentation, especially for time priority and…
The answer explains the micro-price as an expected future price conditioned on order-book information, using imbalance and bid-ask spread as example state variables. Because the price level need not be stationary, the model focuses on increments and tracks…
The document asks whether predicting limit order cancellations has academic or practical value, including for exchange surveillance, spoofing detection, and high-frequency trading. The response emphasizes that spoofing is difficult to define and prove: an…
The document surveys software and approaches for examining limit order books over time. The requested capabilities include displaying bids and asks, switching between individual orders and aggregated price levels, tracking changes through time, replaying…
The document compares simple price proxies for estimating short-term price movement. One proposal weights the best ask price by bid size and the best bid price by ask size; the last traded price is offered as another useful predictor. These measures are…
The document considers whether Least Median Squares (LMS) regression is useful for financial estimation, such as estimating the beta between two securities. LMS minimizes the median of squared residuals rather than their mean, with the intended benefit of…
The document considers forecasting actual high-frequency prices, including bid and ask quotes, rather than forecasting volatility. It highlights practical challenges: model validity may be brief, validity is difficult to assess, and computationally expensive…