คลังความรู้

A trader reports that a futures strategy begins with a position value of negative one and does not act on a later exit signal. The problem appeared after two strategies traded the same contract. Orders were not visible in the strategy’s order callback, and…

VeighNa Fusion is presented as a standardized futures trading platform distributed to customers through partner futures firms. It combines an execution environment, market data tools, CTA strategy development, historical backtesting, and an AI assistant. The…

This discussion explains how to connect a spread-trading strategy with single-contract execution algorithms such as TWAP or best-limit. It distinguishes spread-level execution, which coordinates multiple legs, from algorithms that execute orders for one…

This community post asks how to adapt VeighNa’s double moving-average example for daily bars in live trading and backtesting. The author describes sending ticks to a BarGenerator configured for daily aggregation, binding a daily callback, and loading…

This forum exchange discusses how to use a continuous main-contract series for backtesting while placing live orders in the currently active futures contract. One participant asks whether live use requires building a module to stitch together the continuous…

A forum user asks how to exempt a selected portion of a trading setup from risk control after commenting out a line caused an error. The reply suggests setting the relevant limit higher than the largest order the user expects to submit, so that the rule does…

This support thread discusses errors during futures penetration testing with VeighNa, including error codes 4097 and 4040 and a missing CTPTEST gateway message. The user asks whether the CTPTEST interface must match the futures broker’s CTP API version and…

This forum exchange investigates a VeighNa futures order rejected with error 1007 for an invalid quantity. The logged requests show a one-contract close-today order alongside an open order with zero volume, even though the user expected to submit only a…

This short forum exchange addresses whether a trader needs a historical data service after moving a strategy into live trading. The answer is that initialization may require past market data so the strategy can calculate indicators before processing current…

This forum post asks why a tick-to-bar utility removed a check that discarded ticks timestamped earlier than the most recently processed tick. The author reports seeing this ordering issue in GFEX data for two silver contracts, with incoming tick timestamps…

This Chinese-language forum thread asks whether A-share brokers allow clients to run vn.py on their own servers. The questioner reports that several brokers they contacted required use of a server hosted by the broker. One reply states that, under current…

This support discussion covers login and quote-subscription problems with a simulated market-data connection through a futures broker interface. Participants distinguish market-data login from trading-server login and suggest checking interface versions,…

The post asks why VeighNa’s spread position calculation divides net position by a trading multiplier and rounds down, while an algo’s leg-traded calculation also rounds to the spread’s minimum volume before rounding down. It presents the two code snippets as…

This forum discussion concerns a spread reversal strategy whose two legs sometimes fill together in backtests and sometimes fill days apart. One reply points to a spread-strategy module as a way to address legging. A longer response suggests possible causes:…

This brief community exchange addresses a practical data limitation in VeighNa: a user has a strategy loaded but can access only daily bars, while needing minute bars for a backtest. A respondent suggests two routes: applying for trial access from external…

This tutorial explains how an AlphaModel fits a model from a prepared AlphaDataset and produces prediction scores for a chosen data segment. It describes a shared interface for fitting, prediction, and model details, then distinguishes training, validation,…

A brief forum exchange explains why a trading strategy framework may define parameters as class variables. Doing so lets the strategy engine and user interface inspect parameter metadata before creating a strategy instance, which supports passing values…

The post raises a feature-selection issue in two model implementations: both select dataframe columns by taking everything after the first two columns and before the last. This works when the dataframe is ordered as datetime, symbol, features, label, because…

This brief forum exchange raises a calculation issue in spread trading. A participant describes a spread with a leg ratio of one to three and argues that rounding a leg fill of two contracts after division by three can make the system treat that leg as…

This note describes a futures closing-order workflow using a CTP interface in the SIMNOW simulator. It submits a sell order, polls for its status, and records the submission time. If no volume has traded after a short wait, the example cancels the order and…

The discussion examines why VeighNa’s BarGenerator can produce multi-hour bars whose boundaries depend on when a strategy starts. It attributes the issue to windows being counted from the first hourly bar received, rather than from a fixed exchange-time…

This Chinese-language forum exchange discusses an issue connecting to CITIC Futures’ CTP simulation environment: the market-data and trading servers do not respond after the user starts the connection. Replies suggest first testing login through the…

This forum exchange asks whether live CTP market data is delayed or missing because observed tick arrivals sometimes occur less often than twice per second, and local receipt times differ from the timestamps carried by the ticks. A respondent points out that…

This brief forum exchange distinguishes two kinds of risk control in VeighNa, a trading framework. One participant asks how to force liquidation after a position in an instrument reaches a chosen loss. The reply explains that the framework’s risk controls…