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.
511 documents
The document introduces general and finance-tuned language models, then describes using natural language processing to turn financial text into sentiment measures. It outlines a workflow for collecting and preprocessing Federal Open Market Committee…
The article introduces algorithmic trading as using coded rules to generate and execute orders, then compares it with manual trading. It highlights speed, simultaneous monitoring of markets, reduced reliance on emotional judgment, and the ability to backtest…
This project describes two classifiers intended to predict whether Bank Nifty and its leading constituents would open higher or lower on the following trading day. The stock models use daily OHLCV history and technical indicators for five constituents; the…
The article explains data engineering as the work of collecting, preparing, organizing, and maintaining data so analysts and trading models can use it reliably. It describes engineers as building data infrastructure and pipelines, removing problems such as…
The article considers how increasingly capable artificial intelligence could change trading and financial markets. It distinguishes current rule-based automated trading from systems that learn and adapt, then speculates that AI could assess technical,…
This webinar description explains how high-frequency prices can extend portfolio risk analysis beyond the low-frequency data commonly used in portfolio metrics. The proposed approach uses intraday observations to estimate risk and support portfolio…
This event overview outlines a two-day NSE workshop on algorithmic trading, with material spanning strategy research, trading technology, regulation, and portfolio management. Topics include execution methods such as time- and volume-weighted orders,…
This interview describes David U. Ordiz’s progression from discretionary Bund futures trading to systematic research and portfolio management. His approach focuses on intraday algorithms seeking short-term trend or counter-trend moves across index futures,…
The article introduces multithreading as a way to handle several stock data downloads concurrently. Since network requests spend time waiting for external responses, separate threads can work on different tickers while other requests are pending. It outlines…
The article explains Linear Discriminant Analysis (LDA) as a supervised method for classifying observations and estimating the probability of belonging to a class. It contrasts LDA with logistic regression and describes LDA’s use of Bayes’ theorem, class…
This article introduces FIX as a standardized messaging protocol used to connect participants and systems across electronic trading workflows. It describes how a shared format can reduce integration effort, simplify communication with multiple brokers, and…
This article uses simple betting examples to explain expected value as the probability-weighted average of gains and losses. It shows how a favorable payoff structure can produce positive expectation even when a win is uncertain, while a symmetric…
This document explains ADDM, a method for detecting changes in a trading model’s prediction errors and adapting the model when market conditions shift. Its detector uses a Self-Exciting Threshold Autoregressive (SETAR) model to divide error behavior into…
This interview with trader Priyanka S. includes practical advice for developing and testing equity signals. She cautions that familiar technical indicators such as moving average crossovers may contain little information about future prices, and encourages…
The article explains how a time-series generative adversarial network can produce synthetic financial observations when historical data is limited. It describes the generator and discriminator conceptually, then focuses on the conditional probabilistic…
This project tests a simple ETF pairs strategy in oil, technology, and financial sectors: USO with XLE, XLK with IYW, and XLF with PSCF. It estimates a hedge ratio by regression, evaluates spread stationarity with an Augmented Dickey-Fuller test, then enters…
The document explains how to stitch successive futures contracts into a longer time series for analysis when each individual contract has limited history. Simply joining contract prices can create artificial jumps because adjacent expiries may trade at…
The article introduces spread trading as a hedged position that buys and sells related contracts, such as options on the same security with different strikes or expiries, or futures with different delivery months, commodities, or locations. It recommends…
The article demonstrates simple and multiple linear regression on historical returns for Coca-Cola, PepsiCo, the S&P 500 ETF, and the US Dollar Index. It first uses pairwise correlations, then fits a single-predictor model for Coca-Cola returns using the S&P…
The article introduces the Kalman filter as a recursive method for estimating a changing, partly unobserved state by combining model predictions with noisy measurements and their uncertainty. It explains concepts including normal distributions, variance,…
The article explains divergence as a mismatch between an asset’s price swings and an indicator or oscillator’s swings. It distinguishes regular bullish and bearish divergence, which may warn of a trend reversal, from hidden bullish and bearish divergence,…
This overview explains high-frequency trading as automated order placement that depends on rapid market data, fast decision systems, and low-latency execution. It describes co-location, tick-by-tick feeds, and market making, where firms quote both sides and…
This guide describes a walk-forward workflow for forecasting stock prices with XGBoost. It motivates repeated model updates as a response to concept drift and changing data distributions. Historical price data are cleaned, adjusted prices are used, and…
This project tests a market-neutral pairs strategy on Brazilian equities, grouping stocks by sector and screening pairs with the Johansen cointegration test. It keeps pairs with a consistently signed spread and a half-life no longer than 60 days. Entry and…