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.
682 documents
The post concerns a user who deployed a TradingView signal bot for Binance Bitcoin perpetual futures and encountered an error because the contract setting was empty. The included example robot distinguishes spot from futures exchanges by their names. For…
This beginner guide walks through setting up a Binance account and futures access, creating API credentials, funding the account, and moving USDT into the futures wallet. It then covers registering with FMZ, adding the exchange credentials, renting a…
The article describes a cash-and-carry style trade using spot cryptocurrency and a short crypto-margined delivery futures contract. It monitors the futures premium over spot, buys spot, and opens a matching short to capture the basis; the legs can be closed…
This essay argues that trading models inevitably weaken as market conditions and participant behavior change. It recommends grounding a model in a durable market premise rather than a temporary pattern: for example, a strategy designed for narrow-range…
The document proposes a trend-oriented grid as a position management and exit framework that can be paired with different entry signals. It divides capital into ten portions, opens with one portion after an entry condition, and sets a midpoint, stop, and…
This Chinese-language post describes a stock-selection screen for Shanghai-listed shares. It combines three conditions: daily high-to-low range above one percent, at least one limit-up event within the prior 25 trading days, and a stock code beginning with…
The document describes ten ways a trading strategy’s backtest can misrepresent performance: look-ahead pricing, future data leakage, omitted fees and slippage, unsuitable index contracts, overfitting, limited market capacity, too few observations or signals,…
This tutorial outlines several ways to refine a trading strategy before live deployment. It notes that a signal based on a completed bar may only be acted on at the next bar’s open, while using real-time prices can allow faster entry when a signal appears.…
This introductory overview explains ten general-purpose algorithms and problem-solving methods: quicksort, heapsort, merge sort, binary search, BFPRT selection, depth-first and breadth-first search, Dijkstra’s shortest paths, dynamic programming, and naive…
This article uses the history of Gauss and the normal distribution to introduce a statistical idea relevant to risk management: repeated measurements vary, but their distribution around a central value can help assess typical outcomes and measurement…
This post investigates why a Parabolic SAR calculation appears unchanged when supplied with Heikin-Ashi candles instead of the original price records. It shows JavaScript code that derives Heikin-Ashi open, high, low, and close values from regular candle…
This introductory chapter compares open-source and commercial software for quantitative trading. It describes open-source tools as flexible and inspectable, with the trade-off that users generally need programming skills and may have to solve technical…
The article presents correlation as a guide to building a diversified trading portfolio. For trend following, it recommends combining instruments with lower correlations so positions do not behave like a single exposure. For pairs trading, it suggests…
This explanation contrasts simulated tick-level backtests with live-data backtests. The platform described runs strategies through a repeatedly polled control flow and presents API data as if it arrived during operation. In simulated mode, it interpolates…
This document describes a trading setup that connects TradingView alerts to an FMZ robot and routes commands to spot or futures markets. The example distinguishes long, short, and position-closing actions, and uses exchange direction settings to handle…
The author recounts a path from manual crypto trading and losses to building strategies with FMZ, a trading platform. Experiments included an exchange funding rate strategy, studying high frequency trading, and trying a Martingale approach. A sharp market…
The document explains rounded-bottom, or saucer, formations as either a reversal after a decline or a pullback within an existing uptrend. In both cases, selling pressure appears to ease as price flattens and then turns upward. For a reversal, the pattern is…
This stock screen selects shares with a daily price range above 1%, free float no greater than 5.5 billion shares, and turnover between 2% and 9%. The article presents the combination as a way to find stocks with notable price movement, a smaller share base,…
This article introduces event-driven backtesting as an alternative to vectorized simulation, focusing on how an event loop processes market data and trading actions in sequence. It outlines a queue-based architecture with events for market updates, signals,…
This tutorial explains how Pine scripts execute on FMZ, a platform that supports Pine-style strategies, charting, backtesting, and live trading. It outlines platform differences from TradingView, including settings for execution timing and order precision,…
The article describes using a deep neural network to estimate risk for an automated, beta-neutral long–short equity strategy. Financial statement data and historical prices serve as inputs; the model predicts return quantiles over horizons from five to 90…
A trader reports that a crypto derivatives backtest accepts its first order but rejects later buys, sells, and position closes with a generic parameter error. Their investigation identifies an order price that was too small as one cause, then raises…
This document describes a semi-automated workflow for manually hedging cryptocurrency futures against spot positions across multiple symbols. It tracks both venues' quotes and account positions, displays the futures–spot spread, and provides controls to open…
This overview organizes machine learning methods in two ways: by how they learn from data and by similarities in their model structures. It distinguishes supervised learning with labeled outcomes, unsupervised learning that finds structure in unlabeled data,…