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
This overview contrasts efficient markets, where prices are assumed to incorporate available information, with inefficient markets, where prices may diverge from estimated value. It outlines the weak, semi-strong, and strong forms of the efficient market…
This indicator combines unusual-volume marking with a simplified volume profile. For the time-based view, it compares each bar’s volume with a rolling standard-deviation measure and uses bubble size to indicate the resulting magnitude; candle direction…
This research summary examines factor weighting, orthogonalization, and timing through the relationship between composite information-coefficient weighting and Fama–MacBeth regressions. Under the stated condition that returns and factors are z-scored, it…
The article introduces cross-period hedging by pairing a short position in one contract with a long position in another. It derives how the difference between the opening and closing spreads determines the combined theoretical profit, assuming the stated…
The document describes a mathematical library function for calculating Pearson correlation values between an input signal series and a supplied pattern. It presents the function as producing an array of coefficients, with each value reflecting a comparison…
This Chinese stock-selection proposal screens for daily price ranges above 1%, excludes Beijing-listed shares and certain boards, and ranks remaining stocks by capital-flow strength. The article’s later version describes selecting stocks in the upper half of…
Root mean square (RMS), also called the quadratic mean, is computed by squaring observations, averaging those squares, then taking the square root. The note introduces RMS as a standalone calculation and discusses applying it to a rolling set of…
This overview explains machine learning as a way to model patterns in data and describes how that capability may be applied to investment decisions. It outlines a workflow spanning data collection, feature creation, preprocessing, model training, model…
This indicator adapts a stochastic oscillator’s lookback window to an estimated market cycle. It first smooths the midpoint of each bar’s high and low, then derives in-phase and quadrature components using a digital filter. A homodyne discriminator estimates…
The document explains a simple oscillator formed by comparing a linear-weighted moving average with a simple moving average over the same period. It defines the output as the weighted average divided by the simple average, minus one, and displays the…
This indicator draws a price-chart line by applying polynomial interpolation between two node lines. Its purpose is to create a smoothed or shaped path from the selected price series, with the nodes set at a chosen distance in bars. The available controls…
This indicator seeks potential market tops and bottoms by comparing short and long Laguerre-smoothed price series. It calculates a percentage price oscillator from their difference, then ranks the current oscillator against its values over a rolling lookback…
This 2020 market note reviews Chinese equity style performance and outlines a short-term outlook. It reports that small-cap reversal strengthened, value began to recover, and earnings-growth exposure showed a brief rebound. Its recommendations draw on…
The document presents an indicator-processing method that smooths the Bears Power time series by filtering higher-order harmonics in its spectrum. It suggests that the same approach can be applied to other indicator outputs. The stated advantage is…
This document describes a data preparation workflow for Malaysian crop production, harvested area, producer prices, land use, population, and GDP data over multiple years. It joins production with harvested area to calculate inverse yield, organizes crop…
This document describes an extended fractal indicator that lets users set how many candles to examine on each side of a potential turning point. It also includes a minimum-distance input for distinguishing the extreme point from adjacent peaks. Separate…
This document presents a simple indicator for measuring how far the current closing price sits, in percentage terms, from an exponential moving average. It calculates the average over a configurable period, then expresses the difference between the close and…
This article summarizes a study of U.S. mutual funds that links SEC N-SAR options disclosures with fund returns and characteristics. The sample covers 1998 through November 2013 and includes 106,357 filings. Cross-sectional and panel analyses compare funds…
The article explains why standard stochastic gradient descent can remain inefficient when training Word2Vec on a large vocabulary. Even when an update concerns only a few words, calculating gradients across the full parameter matrices wastes work, and the…
ColorXdinMA_StDev is a moving-average indicator that adds colored dots to communicate trend direction and relative trend strength. It measures the standard deviation of the ColorXdinMA indicator and compares that value with two configurable thresholds. When…
The document summarizes research into whether changes in Google search activity for finance-related terms can help explain stock-market fluctuations. It frames search volume as behavioral data: shifts in what people seek online may offer insight into…
This brief support exchange addresses whether a strategy’s data is standardized anew each day during simulated trading. The response says that when the strategy uses parameters bound for live trading, simulated trading standardizes that day’s data. A…
This document explains an indicator that converts each bar’s timestamp into hour-of-day features for data collection and modeling. It provides 24 binary buffers, one for each hour, and an additional buffer containing the numeric hour. For each bar, the…
This overview explains XGBoost regression as an ensemble of decision trees built sequentially to reduce a loss. Each new tree uses gradient information from prior predictions, while complexity controls and regularization are intended to limit overfitting.…