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 discusses how to estimate hedging costs for Chinese equity index futures. It argues that raw futures premiums or discounts need adjustment for time to expiry, convergence, and expected dividends. A dividend model and quadratic equation are…
VolatilityMA is a moving average intended to help identify trade direction. It calculates an average of a chosen price series while adapting the averaging length to changes in normalized price variability. The description says the lookback resets to its…
This indicator converts closing prices into rolling standardized deviations from their recent averages, then smooths the results across three lookback horizons. The supplied periods are 180, 60, and 20 observations. Each series subtracts its lookback average…
The document explains how to measure Expert Advisor runtime in MetaTrader 5 while excluding delays that can distort the tester’s reported duration. It distinguishes total run time, which can include history synchronization and tick preprocessing, from net…
This stock-selection rule screens for Chinese equities with a 14-period RSI below 65, a circulating market value between 50 and 100 billion, and a current low below the previous day's low. The accompanying explanation presents RSI as a gauge of overbought or…
This overview explains how deep learning models can be assembled from input, intermediate, and output layers in a visual strategy-building platform. It surveys layer families including convolution, pooling, recurrent networks, embeddings, noise and dropout,…
This indicator description presents GRFLeadingEdge as an alternative to Bollinger Bands intended to reduce delays associated with averaging. It says the method resembles a standard-deviation channel built by smoothing a straight line over a fixed number of…
This research review examines whether predictable stock returns in particular months reflect changing risk or temporary mispricing. Its central test is seasonal reversal: if a stock earns unusually high returns in one calendar period because demand…
This analysis compares limit-up activity in Chinese equities around the implementation and anticipation of registration-based reforms. It reports that daily limit-up counts fell in the later part of the observed half-year, with lower five-day averages and no…
This implementation explains how a bivariate Gaussian copula represents dependence between two variables after their observations have been converted to uniform pseudo-observations. It estimates the dependence parameter by mapping those observations through…
The article proposes reducing noise in stock prices with wavelet decomposition, fitting ARMA models to the resulting coefficient series, forecasting those coefficients, and reconstructing a one-step-ahead price estimate. Its denoising explanation relies on…
This document compares several common rate-of-change measures: raw momentum, percentage change, and ratio forms. It gives the corresponding formulas and maps several variants to names used in TA-Lib, TradeStation, and MetaStock. These definitions help…
The document describes a Kalman Filter indicator intended to smooth price noise and extract an underlying trend. It distinguishes this recursive approach from filters that rely on a history of observations: each update uses the prior state estimate and its…
The document introduces the Psychological Index as an overbought and oversold indicator, noting that it was described in Futures Magazine in June 2000. Its calculation starts by marking each period as an up day when the close is higher than the preceding…
The FinRL data layer is presented as a unified processor for accessing market data from multiple APIs, cleaning it, and extracting features. Users specify a date range, stock list, interval, and other parameters. The document distinguishes missing…
This discussion explains a mismatch in which a simulated trading run produces no signal even though a backtest does. The reported cause is a SQL query using a one-row lead on closing prices. At date t, that field requires the closing price from t+1, which is…
The DZP Trend indicator aims to show price movement after accounting for the underlying trend. It compares the percentage change in the applied price over a chosen shift with the percentage change in an exponential moving average over the same shift. The…
This post describes a revised rolling machine-learning training workflow, reporting that its code was reorganized for clarity, model parameters were adjusted, and memory monitoring was added. The author says the parameter changes increased backtest speed…
This indicator applies the Ehlers Fisher Transform to RSI calculated from a moving average of price. The transformation is intended to normalize RSI values and make changes in market conditions easier to see. Price smoothing can use simple, exponential,…
The document explains how to generate moving-average features in BigQuant when only selected lookback windows are wanted. It contrasts a range-based list comprehension, which the platform accepts, with a list of chosen values, which it says is unsupported in…
This guide outlines a workflow for preparing a factor research report for a quantitative trading competition. It recommends generating factor data, then using BigQuant’s FactorLens v4 to calculate single-factor results. The platform returns ranking metrics…
This text introduces market efficiency as a contested idea whose meaning shapes how investors approach investing and valuation. It says the chapter offers a basic definition and considers what efficient markets would imply for investors. It also points…
This indicator studies whether price tends to rise or fall on the bar after a run of consecutive candles with the same direction. It measures runs from one to twelve candles and simulates a long position entered at the close of the run and exited one candle…
The article proposes a Chinese equity screen restricted to the metaverse industry. It looks for stocks opening near or above their ten-day moving average, then ranks candidates using a measure combining the session’s price change with large-order activity.…