Resumos e ideias principais, escritos pelo agente de investigação da Stratmill, dos livros, artigos científicos, artigos e código consultados pelos nossos agentes de IA. Cada página inclui uma ligação para o original.
560 documentos
A user asks whether vn.py's backtesting interface, which accepts a single instrument code, limits backtests to one stock or futures contract at a time. The reply points to the Jupyter portfolio backtesting example as the route for testing a portfolio of…
This troubleshooting post examines why position records from an options gateway show zero unrealized profit and loss. The author traces the issue to the gateway’s position-query handler: it initializes average price at zero, while the received…
The document describes a question about order conversion in a trading strategy. When a close order is submitted, the converter calculates frozen volume from the unfilled order quantity. The author says this calculation sets a short-side frozen amount, which…
The post discusses whether a target-position strategy in VeighNa’s CTA framework submits adjustment orders one bar late when running a bar-based backtest. It shows a code change that sends a new order when an earlier order is still unfinished, specifically…
A VeighNa forum exchange distinguishes stopping live trading from stopping a strategy’s market-data processing. After initialization, a strategy can remain subscribed to quotes and continue calculating or printing tick-based signals; trading calls do not…
A trader reports that a sell-to-close stop order on VeighNa appeared active, then was marked triggered by the CTA strategy interface but canceled in the vn.py trading interface when the opening auction tick indicated a price below the stop. The position…
This forum exchange explains how to run VeighNa CTA backtests with bar data held in CSV files. One approach converts CSV rows into BarData objects and assigns them directly to the engine’s history data, bypassing database loading. The bars must be sorted…
The discussion answers whether a trading strategy parameter can use a character or string value. One community member replies that it can, indicating that strategy parameters need not be limited to numeric types. This may be useful when a strategy accepts a…
A brief forum exchange discusses why five-minute bars generated from one-minute data may differ from Wind data. A user suspects that one-minute bar alignment or delayed CTP data could explain a bar appearing at 9:00. The response clarifies that VeighNa…
The post describes modifying vn.py’s chart widget so incoming live bars can be combined into larger minute intervals. Its approach maps interval labels to minute counts, compares each incoming bar’s timestamp with the most recent stored bar, and, when the…
This brief forum exchange addresses whether VeighNa’s CTA strategy framework can be used for stock trading and whether its OffsetConverter is needed. The response says that stock strategies can use the framework by placing long and close-long orders without…
This announcement outlines a two-day VeighNa training course on machine-learning cross-sectional multi-factor strategies, also described as alpha strategies. It presents their use in equity selection, index enhancement, absolute-return portfolios, and…
This forum exchange explains why a VeighNa installation on Unraid may not show a web page. VeighNa Station and VeighNa Trader are described as PyQt desktop applications that need a graphical environment; an installation mode with a GUI does not mean the…
A brief forum exchange explains why a contract may remain listed in the graphical interface after a manual close, even when its quantity is zero. The user had changed a position value in a JSON file and asks how to remove the contract from the display. The…
A trader asks why orders in a stock-trading simulator appear to fill at exactly the submitted price, even when the order price is adjusted by several ticks. The trader is concerned that this behavior makes it difficult to judge whether a strategy’s results…
A user asks how to calculate and continuously update a moving average from completed one-minute bars pushed by the Xuntouyan real-time market-data interface in vn.py. They want to use the provider’s already aggregated bars instead of creating bars with…
This short forum exchange explains why a Zhengzhou Commodity Exchange contract may show a successful market data subscription but deliver no ticks through a CTP connection. The example uses soda ash: a trading screen displays SA2601, while the CTP interface…
This short forum exchange asks whether downloaded Ricequant K-line data is stored in the local VeighNa database and whether repeated timestamps could affect backtests. The response says that data saved under the same database path is stored with a unique row…
A forum exchange addresses whether VeighNa supports an interface for an investor with an account at CITIC Securities. One reply says support is available, while noting that the user may need to apply separately for API access. The response also points out…
This short forum exchange asks what happens when a strategy holds one long contract but submits two identical one-contract sell stop orders, while the account also holds long positions opened by other strategies. The question contrasts three possible…
A trader using VeighNa 3.9.1 with the SimNow CTP simulator reports that manual opening and closing orders are canceled, even after stopping strategies and restarting. Algorithmic orders still work, suggesting the issue is specific to the manual order entry…
This forum thread combines a troubleshooting exchange about a futures gateway with code excerpts for mapping a broker API into VeighNa. The code translates broker order-state values into framework statuses and maps direction, order type, open/close offsets,…
This forum exchange explains why a VeighNa strategy may remain outside its trading state during a backtest, leaving it unable to submit orders. The response says the backtest must load more historical data than the number of days requested by the strategy’s…
A trader asks about an unusually large backtest result for an index spread strategy on a day when the underlying market fell sharply. They report that the fills show no leg being left unpaired, so the source of the result is unclear. The discussion does not…