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 career overview explains the work of quantitative traders and the mix of skills commonly needed to research and automate trading strategies. Quants analyze market data, develop models, test ideas, and implement trades using computer programs. The…
This reading note summarizes the KKT business-cycle index, a data-driven measure proposed in a referenced academic paper. The index uses Mahalanobis distance and statistical likelihood to compare current economic conditions with a robust-growth state,…
The discussion clarifies the meaning of a factor-crowding value shown in BigQuant’s factor analysis module. The reply says the value is the recent average turnover rate of stocks after they have been grouped by a factor. It is therefore a group-level average…
This research summary describes using an imitation portfolio to infer the holdings of three Chinese mutual funds. The motivation is that public disclosures arrive with a delay and represent holdings at reporting dates, making them an imperfect guide to a…
The document defines an Average Modified Moving Average indicator with two inputs: a calculation period and the price series to apply. Each new value combines the previous indicator value, weighted by the period minus two, with the current and prior prices,…
This strategy forecasts the future value of a spread between cointegrated assets, then compares the forecast with the current spread to generate trades. The document describes three approaches: trading predicted spread returns directly, following spread…
This Chinese-language post describes an equity screen requiring turnover between 3% and 12%, a 2021 listing year, and rising price lows. It interprets successively higher lows as a sign of improving rebound strength and possible momentum or trend. The post…
This study analyzes information transmission among 28 first-level Chinese equity sectors using daily closing prices from 2000 through 2017. It uses transfer entropy to estimate the strength and direction of information flows between sector time series, then…
This proposed equity screen combines a daily amplitude threshold with a ratio involving the prior day's turnover and current auction volume relative to prior-day volume. It also uses a rising 30-day moving average as a trend filter. The discussion describes…
This short indicator description presents Time Averaged Price as a line calculated from prices over a selected period. It identifies two settings: the calculation period and the applied price. Examples are shown for close, high, low, median, and weighted…
This survey introduces regression methods used in financial modeling, from ordinary linear models to polynomial and logistic regression. It also explains quantile regression for examining different parts of a conditional outcome distribution, including its…
This practitioner note compares very high-return, moderate-return, and lower-return backtests for short-horizon equity models, focusing on whether their apparent performance persists out of sample. The author reports that aggressive timing filters and models…
This discussion contrasts quantitative trading with manual trading, emphasizing that the main difference lies in execution. It characterizes quantitative trading as using mathematical or statistical models and historical data to identify strategies that may…
This post describes a Chinese equity screening rule that ranks stocks by capital inflow strength and limits candidates to a turnover rate between 2% and 9%. It suggests that strong inflows may indicate investor attention and buying interest, while the…
The document describes an indicator concept that searches across moving-average periods and selects one using historical trade expectancy, combining win rate with average reward. It also proposes using maximum favorable excursion to estimate a typical trend…
This document proposes a Chinese equity screen combining amplitude above 1, rising lows, and return on equity above 15% for five consecutive years. The technical conditions aim to capture price movement and a strengthening price floor, while the ROE…
The document explains how NumPy and pandas serve different but complementary roles in Python data work. NumPy provides multidimensional arrays and efficient numerical operations, making it suited to mathematical calculations. pandas builds on NumPy with…
This stock screen selects shares with RSI below 65, circulating market capitalization between 5 billion and 10 billion yuan, and total shares of at least 200 million. Its stated aim is to combine a technical condition with company size and liquidity-related…
This translated reference explains NumPy’s matrix subclass and how its behavior differs from the general ndarray type. It describes matrix construction from strings, nested sequences, and arrays, along with helper functions for creating or combining…
This stock selection rule combines a 14-period RSI below 65, three consecutive declining sessions, and a positive price-to-earnings ratio. The article frames the conditions as a blend of technical behavior and a basic profitability valuation filter, and…
The article explains why a high win rate alone does not ensure long-run profits. Traders may close winners too early and let losing positions grow, so the size of gains and losses matters alongside their probabilities. It also argues that trading systems…
This indicator method looks for divergence by fitting linear regression channels to price and to an oscillator, using RSI as the example. It compares the slopes of the two channels: when they differ, the indicator changes the color of the channel's middle…
This stock-selection proposal filters for amplitude above 1, weekly MACD above zero, and turnover between 2% and 9%. It combines a price-range condition and a weekly trend measure with turnover as a proxy for trading activity and liquidity. The document…
This document introduces the goal of deriving gradients for Word2Vec’s cost function with respect to a target vector’s weights. It says the derivation uses a set of expressions and formulas, then works through the gradient calculation step by step. The…