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.
5,018 documents
The article surveys career paths where people with quantitative, statistical, and software skills can apply machine learning outside traditional asset management. It discusses financial technology use cases such as liquidity prediction, credit assessment,…
This career guidance article outlines academic subjects that support work as a quantitative analyst, researcher, trader, or developer. It assumes a numerate degree background and identifies probability as a core foundation, with measure theory as useful…
The article introduces K-means as a hard clustering method and applies it to daily open, high, low, and close data. It explains that the algorithm assigns each observation to the nearest cluster centroid and iterates assignments and centroid updates to…
The article demonstrates fitting Gaussian hidden Markov models (HMMs) to simulated returns and S&P 500 daily returns. In the simulation, bullish and bearish periods are generated with different means and variances; a two-state model is then fitted with…
This article proposes a staged learning path for people preparing for quantitative trading research. It begins with expectations about the work: systematic trading relies on scientific hypothesis testing, statistical analysis, programming, and…
This article outlines second-year undergraduate mathematics topics for self-study, with links between foundational theory and quantitative finance. Its roadmap includes the Riemann integral, metric spaces, vector calculus, nonlinear ordinary differential…
Unsupervised learning finds structure in observations without labeled outcomes. The article contrasts this with supervised prediction and explains that unlabeled methods are useful when labels are costly or data is high-dimensional. In finance, possible uses…
This article outlines an MQL5 Expert Advisor that combines a linear regression forecast with technical indicators and gradient descent. Its proposed objective is to reduce absolute forecast error by adjusting model coefficients; the learning rate controls…
This article is a curated directory for traders beginning to study machine learning in algorithmic trading. It organizes suggested learning materials into books, online courses, videos, blogs, interviews, and scientific papers. The topics represented include…
This article introduces an MQL5 library for specifying, fitting, forecasting, and assessing conditional volatility models. Its architecture separates the conditional mean, volatility process, and residual distribution, with a mean-model class serving as the…
The article introduces a pipeline for applying topological data analysis to financial time series. Takens time-delay embedding converts a one-dimensional series into vectors in a higher-dimensional point cloud, controlled by embedding dimension and delay.…
This article presents an MQL5 Expert Advisor that combines UK gilt and FTSE 100 market data in a linear regression model. It describes conventional gilts and inflation-linked gilts, and outlines the common tendency for investors to move between equities and…
This article extends a context-aware reinforcement learning framework for cryptocurrency trading with a risk management module. In the described two-stage setup, market conditions are classified by trend and volatility so specialized sub-agents can learn for…
The article explains how offline reinforcement learning can become overconfident when its training data do not cover the states and transitions encountered later. It introduces SPLT-Transformer, which uses separate Transformer-based latent models for policy…
This article shows how to calculate technical indicators directly from input vectors in MQL5, giving developers control over the data series supplied to each calculation. It contrasts this approach with built-in indicator handles, which use predefined price…
The article diagnoses why sequential-bootstrap and standard-bootstrap models produced nearly identical out-of-fold Brier scores. It traces the problem to two linked implementation defects: a pipeline helper replaced the sequential bagging classifier with a…
This article presents an MQL5 money-management module that adjusts position size as the recent price range expands. It uses a monotonic queue to track rolling highs and lows, describing how each observation is added and removed at most once to achieve linear…
This article outlines a workflow for building a regression model to forecast an asset’s closing price within a week and connecting it to the MetaTrader 5 Strategy Tester. It compares linear, polynomial, decision tree, and support vector regression,…
The article adapts ideas from Molformer, a transformer framework for molecular graphs, to represent financial data. Its proposed market representation combines nodes for individual assets with nodes for recurring multi-asset patterns, aiming to encode…
This technical overview explains how activation functions map neural network inputs to outputs and why their derivatives matter for backpropagation. It presents formulas and MQL5 usage examples for functions including ELU, exponential, GELU, sigmoid, tanh,…
The article describes an MQL5 entry signal that encodes each of the last 64 bars as a binary value: bullish bars set a bit, while other bars leave it unset. A perceptron then adds the weights for active bits to a bias to produce an activation score. The…
This article introduces classifier ensembles and explains bagging, or bootstrap aggregation, as a parallel method for reducing prediction variance. Each base classifier is trained on a bootstrap sample, and their outputs are combined through averaging or…
This article explains convolutional neural networks as a way to recognize patterns in price-chart data. It describes how convolution filters scan input sequences to produce feature maps, while subsampling retains local maxima or averages to reduce…
This article completes a trading-oriented implementation of RefMask3D, a language-guided 3D point-cloud segmentation framework. It focuses on the Object Cluster Module, which combines two self-attention blocks, an intervening cross-attention block, and a…