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 document describes an oscillator intended to show the balance between bullish and bearish price pressure. It defines separate bull-power and bear-power inputs, applies a moving average to each, and says the calculation uses configurable periods and…
This document summarizes a research report on estimating the duration of public bond funds. It describes two approaches that use information from different reporting streams: portfolio holdings disclosed in quarterly fund reports and fund net asset values…
This discussion asks why a displayed time may reflect the hosting machine’s clock and how to obtain local time instead. It also distinguishes a human-readable date string from a numeric timestamp, which can be useful when a program needs to store a time…
The document summarizes research on forecasting the correlation between stock and bond returns, a quantity used in portfolio construction, hedging, and risk assessment. It argues that simply extrapolating historical monthly correlations can be unreliable…
The code implements a time-series prediction model paired with a TRA component that can combine predictions across multiple states. A base model, such as an LSTM, produces hidden representations and state-specific predictions. When multiple states are…
The report describes two steps for improving an equity alpha model: remove factors that duplicate information, then optimize the remaining signals while accounting for their correlations. It uses Fama–MacBeth regressions to screen factors and reports…
This indicator calculates volume-weighted average price over weekly, monthly, and yearly periods. For each bar, it divides the sum of typical price multiplied by volume since the relevant period began by the corresponding cumulative volume. It then plots the…
This note outlines an equity screen for companies associated with beverage or alcohol imports and exports. It selects stocks with RSI below 65 and a circulating market value between 5 billion and 10 billion yuan, then uses the company’s business area as an…
The Hurst exponent describes how strongly observations in a time series remain related across long lags. The document explains its origins in hydrology, where it was developed to study long-run variation in river flow, and connects it to the decay of…
This Chinese equity screen combines turnover between three and twelve percent with a reversal-candle condition and a company-status filter. The example formula also requires a low price below a stated threshold, a candle ratio measure at or below 0.2, and…
This forum discussion explains an apparent five-minute discrepancy between bars produced by different charting and trading systems. One participant observes that some platforms label a bar with the time at its end, while the discussed VeighNa bar appears to…
The Balance of Power indicator estimates whether buyers or sellers controlled a candle by comparing the close with the open and scaling that difference by the candle’s high-low range. Positive readings suggest buyer dominance, negative readings suggest…
This study evaluates analyst-consensus valuation and earnings factors for stock selection across the Chinese equity market and its industries. It compares factor information coverage and effectiveness, using information coefficients to assess the…
The Volume Accelerated Directional Energy Ratio combines normalized price direction with volume to estimate the strength of buying and selling pressure. Price change is scaled by a recent high-low range and bounded, then multiplied by volume in one of three…
The document describes a stock selection screen requiring turnover between stated lower and upper bounds, first-level bid volume above ask volume, and a ratio of buyer-initiated to seller-initiated trading above a specified threshold. It presents these…
This short description explains a step chart indicator that ignores price moves smaller than a chosen pip threshold. By suppressing changes below that threshold, the chart represents price as a sequence of steps and filters out smaller fluctuations. The…
This report applies a parametric portfolio policy approach to equity factor timing, using external variables to adjust factor weights. It tests the approach on Chinese A-share data and compares two ways to use information: time-series models and…
This report describes a Chinese equity selection factor called CPV, built from the relationship between intraday prices and trading volume. It combines three characteristics of daily price-volume correlation: its average level, variability, and trend. The…
XDPO_Histogram is described as a simple oscillator built from the difference between closing price and its moving average. That difference is averaged again over a smoothing period, and the resulting oscillator is shown as a color histogram. The calculation…
This code collection provides helper functions commonly used when building quantitative factors from time-series data. It wraps TA-Lib operations for rolling sums, standard deviations, minima, maxima, momentum differences, and weighted moving averages. It…
This proposed Chinese stock screen combines daily price amplitude above 1, a listing history longer than three years, and revenue for 2021 exceeding 2018 revenue by a ratio greater than 1.1. The article presents the revenue comparison as a basic growth…
This indicator identifies five-point Merrill M and W wave shapes from turning points found by a modified ZigZag process. It plots the patterns and support or resistance levels at extreme points, then estimates possible price movement based on how patterns…
This brief description introduces a simple moving average calculated using a geometric mean. It identifies two configurable inputs: the number of observations in the calculation period and the price series used as the input. The indicator therefore differs…
The proposed equity screen combines three filters: membership in a metaverse-related theme, a prior-day share price above its 250-day moving average, and top-of-book bid volume greater than ask volume. The document interprets the moving-average condition as…