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.
21,023 documents
The study defines an earnings surprise by comparing reported results from forecasts, flash reports, or periodic filings with analyst consensus net profit estimates from the day before the announcement. For nonannual reports, it estimates full-period…
This brief literature note introduces a VaR-augmented Black-Litterman approach for constructing an absolute-return fund-of-funds portfolio with market-risk controls. The described formulation incorporates value-at-risk alongside practical trading…
This research review examines whether mutual-fund outflows provide a clean measure of nonfundamental pressure on stock prices. It decomposes the MFFlow measure into components tied to the inverse of returns, the inverse of turnover, and relative fund selling…
The document explains lookahead bias: a backtest can accidentally use future candle data because the full historical dataframe is loaded before indicators and signals are calculated. This can make results appear unrealistically strong. It describes an…
This Chinese-language report summary reviews the history and practice of machine learning in quantitative investing. It says that machine learning was used in the field during the early-1990s boom and that its applications continued in algorithmic trading…
This report overview describes the longstanding use of machine learning and artificial intelligence in quantitative investing. It notes that applications were already present during an early-1990s wave of interest, and that use continued in areas such as…
This indicator estimates buying and selling pressure from each candle’s range around its open. Buying pressure is defined as the high minus the open, while selling pressure is the open minus the low. Prevailing pressure is set to whichever of those two…
The document introduces multiple linear regression as a model relating one dependent variable to several explanatory variables, contrasting it with simple regression. It connects the method’s broad use in quantitative finance with the progression from the…
The document describes a software implementation of the Johansen cointegration method for forming mean-reverting portfolios from asset prices. It computes cointegration vectors, orders them by eigenvalue, and converts each vector into hedge ratios normalized…
The document explains TWR as a candle-reconstruction indicator: it uses the previous bar’s color and the current bar’s close while disregarding other price information, then redraws the candle chart around closing prices. In its color convention, red marks a…
The document describes a tool for checking whether a trader’s actions and recent outcomes are associated with subsequent losses. It trains a neural network on eight features from closed trades, then reports validation accuracy relative to a majority-class…
This stock screen combines a turnover-rate range of 3% to 12% with a reversal condition and a pattern labeled a Morning Star. The accompanying indicator logic adds a large intraday range relative to the previous close, declining short-, medium-, and…
The document describes John Ehlers’s recursive median oscillator, which combines a median filter with recursive smoothing and a high-pass filter. The median of a short window of closing prices reduces the influence of extreme observations that can distort…
This brief request proposes a factor based on the correlation between a benchmark’s returns and an individual stock’s returns over a three-day window. It names the intended inputs and lookback, but provides no code, worked example, explanation of how to…
This April 2020 market note reviews a sharp decline in Chinese equities amid weakening fundamentals and falling investor sentiment. It reports that large- and small-cap indexes had similar declines, while consumer and utility sectors held up better and…
This note explains how to handle low-frequency financial or analyst factors alongside daily price and volume factors when constructing nonlinear models. One approach is to merge monthly or quarterly fundamentals with daily market data and carry the latest…
This brief note describes a feature-selection exercise based on a random forest template. The author says they ranked 37 stock factors by random forest importance and retained the nine factors with the highest importance scores. The document does not…
The document explains how to find the highest and lowest closing prices over roughly two years using rolling time-series maximum and minimum functions. It treats two years as about 400 trading days and applies those functions to the closing-price series,…
The document explains the Generalized DEMA (GD) moving average, which combines a price’s exponential moving average with a second exponential smoothing of that average. A volume factor controls the secondary smoothing contribution, while the period and…
This research summary proposes measuring capital flows between large-cap and small-cap styles to identify when style investors are shifting allocations. It treats flow intensity as a signal of changing conviction and risk: when intensity is high, investors…
The article challenges three barriers often thought to exclude individual investors from quantitative trading: advanced mathematics, coding skills, and large capital. It presents quant methods primarily as statistical tools for understanding markets, and…
This dashboard description explains how Pearson correlations across symbols and timeframes can support trading decisions. A color-coded matrix is intended to show positive and negative relationships, with selectable daily and intraday views. The examples…
This ProRealTime indicator fits a least-squares line to recent closing prices, using a fixed lookback of 200 bars. It calculates the slope and intercept from sums of price and time values, then draws the fitted line and bands around it. The described band…
This brief BigQuant forum exchange concerns errors in a factor-analysis template when a factor’s coverage falls below the platform’s threshold. One concrete cause is division by zero: the expression using the difference between the high and close prices can…