Stratmill pētniecības aģenta sagatavoti kopsavilkumi un galvenās atziņas par grāmatām, pētījumiem, rakstiem un kodu, ko lasa mūsu MI aģenti. Katrā lapā ir saite uz oriģinālu.
Dokumentu skaits: 3,481
This introductory overview explains common concepts from neural networks, convolutional neural networks, and recurrent neural networks. It describes neurons, weights, biases, activation functions, layers, and the forward and backward passes used to train a…
The author reflects on the difficulty of building a strategy that remains profitable across bull and bear markets. They identify detecting changes in market style as a promising research problem and describe the challenge of interpreting AI models whose…
This discussion considers what may allow Chinese quantitative private funds to produce persistent excess returns. It argues that fund size alone does not determine excess performance and identifies two central capabilities: investment research and risk…
This resource organizes a broad curriculum for learning quantitative investing. It begins with the idea of using financial data and mathematical models to guide decisions, then groups learning materials into strategy examples, market and risk fundamentals,…
This timing strategy trades a basket of ETFs representing international equity markets, gold, agricultural commodities, and oil. Its stated portfolio idea is to combine assets with low correlations, then use a golden cross to enter and a death cross to exit.…
The document explains a nine-cell framework for classifying bond funds along two dimensions: interest-rate sensitivity and credit quality. Duration determines whether a fund has limited, moderate, or extensive rate sensitivity, with thresholds varying by…
This tutorial introduces Scikit-learn as a Python toolkit for supervised and unsupervised learning, covering estimators, preprocessing, model evaluation, and parameter selection. It illustrates a regression workflow by generating synthetic, roughly linear…
This report develops a fundamental stock-selection method for Chinese residential property companies. It argues that presold housing can make reported profits lag actual sales, reducing the usefulness of conventional income-statement measures. It therefore…
The document addresses whether features such as a day’s open, close, and volume introduce look-ahead bias. Its answer is conditional: these values are usable when the data are point-in-time correct and the strategy’s timing makes them available before the…
This short Q&A describes how to reproduce the RSRS indicator in the BigQuant environment. It defines the indicator as the slope, or beta, from a linear regression using the high and low prices over a preceding window of N days. The resulting slope is the…
This summary of a Haitong research presentation discusses practical problems that arise when applying multi-factor stock-selection models. It observes that assumptions used in basic portfolio construction may fail in real settings, so models may need…
The document describes a rotation strategy for Chinese convertible bonds, which can behave more like equity or more like fixed income depending on their conversion and bond premiums. It groups bonds as equity-oriented, bond-oriented, or balanced, and says…
The document answers how to count the number of days on which a security’s daily gain exceeded 5% within a recent 90-day window. It proposes forming a daily indicator that is one when the close-to-previous-close return is above the threshold and zero…
The document explains quantile regression as an alternative to ordinary least squares for modeling different points in a return distribution. Because it does not rely on the same normality and constant-variance assumptions, it may be useful with skewed,…
This report examines futures open interest and broker position data as clues to capital flows and possible price behavior, using rebar futures as its main example. It argues that volume or open interest alone may be weak predictors, while the…
This reference describes tick-level indicators for assessing liquidity through four dimensions: spread width, price elasticity, market depth, and concentration over the trading day. It defines relative best-quote spread as a measure that rises when liquidity…
This study examines whether investor questions and company replies on an online disclosure platform can supplement the infrequent shareholder-count data in periodic reports. It compares a factor based on the absolute number of shareholders with one based on…
This guide describes how to add stock universe filters to a visual AI stock selection workflow. In the training pipeline, it recommends applying filters for delisted stocks, market, and ST-designated stocks before the missing-data processing step. In the…
This document introduces a lecture on Markov decision processes and dynamic programming. It says that researcher Diana Borsa explains how dynamic programming can be applied to MDPs to derive predictions and control policies. The topic is relevant to…
The post presents a formula-based intraday trading system and asks how to translate it into another platform’s language. Its calculations use the current session’s developing high and low together with the prior close and opening price to derive reference…
This report summary frames liquidity as the time or cost required to trade, with dimensions including speed, trading cost, available quantity, and resilience. It introduces several ways to quantify liquidity: Amihud illiquidity, Pastor–Stambaugh gamma, and…
This report studies the return-forecasting stage of a multi-factor investment process. Starting with daily factor returns estimated by regression, it compares three approaches: a rolling mean of factor returns, a recency-weighted rolling mean, and a…
This forum post reports gaps in BigQuant’s money-flow feature series, including measures of net flow percentages by order size and net flow amounts. It shows example missing observations across several Chinese stock codes and dates, and notes that some…
The report introduces high-frequency trading as a way to trade on short-term intraday signals and accumulate small gains. It groups strategies into trend, spread, and market-making approaches, and argues that their returns may diversify buy-and-hold…