天马量化

TECHNICAL LOGIC

Multi-factor selection + multi-strategy fusion
driving a dynamic quant system

ALPHA SOURCE

Where our alpha comes from: a dynamic multi-strategy engine

Tianma Quant is not a single strategy. It is a fully automated system driven by multi-factor selection and multi-strategy fusion.

Market data → Factor screening → Strategy matching → Rotation → Alpha

ENGINE

Three parts of the engine

01

Factor library

A library of mature alpha factors built into the system.

  • Momentum
  • Value
  • Growth
  • Volatility control
  • Proprietary composite

02

Adaptive matching

A multi-model, self-adapting strategy pool.

  • AI-based matching
  • Best strategy chosen per instrument
  • Adapts to regime and instrument character
  • Personalised configuration

03

Automatic filtering and rotation

Dynamic factor screening and automatic multi-strategy rotation.

  • Scores every strategy and instrument
  • Drops what stops working
  • Rotates into better models in real time
  • Optimises allocation

Through this engine the system keeps re-optimising itself as markets change.

RISK CONTROL

Risk control in three layers

Tianma AI runs a three-layer risk framework to protect capital and keep returns steady.

01

Portfolio diversification

Each account is split into equal slices — CNY 1m across ten names, for example.

    Goal: a single blow-up costs one slice, not the account. Non-systematic risk is spread thin.

    02

    Dynamic cash reserve

    Part of the capital is held back to absorb losses after stop-outs.

      Goal: the starting position size never changes — what varies is the percentage, not the absolute capital.

      03

      Cyclical trading

      Trading runs in cycles, typically a year, inside a defined universe.

        Goal: capture the year's key moves and avoid being fully invested in a bad tape.

        PIPELINE

        End-to-end automation

        From validation to live trading — every step can be reviewed, compared and reproduced.

        01

        Test

        • Built in: three complete strategy products
        • How: back-test any strategy against any instrument
        • Output: returns computed on historical data

        Transparent evaluation of long-run behaviour.

        02

        Simulate

        • Built in: run a strategy on a simulated account
        • How: trade against current conditions
        • Feedback: watch the strategy react in real time

        A risk-free rehearsal in live conditions.

        03

        Go live

        • Built in: connect your existing brokerage account
        • How: pick a strategy and instruments, then run
        • Edge: no manual intervention — orders execute automatically

        Execution is handed to the system.

        The value: an automated pipeline lets anyone pick up quantitative trading quickly — from strategy validation to simulation to live trading, every step is designed to be simple and visual.

        ADVANTAGE

        Four core advantages

        Tianma AI brings four core advantages to the table.

        01

        Fully automated AI strategy

        A programmatic system that removes the limits of manual trading — faster decisions, more precise execution.

          02

          Direct broker VIP access

          Connected to VIP execution channels: lower order latency, steadier connectivity, higher fill rates.

            03

            Discipline over human weakness

            Signals are executed as written. Built-in stop-loss and take-profit lock in gains and cap drawdowns.

              04

              24/7 monitoring

              The system watches the market continuously and enters or exits at the right moment, unattended.

                So investors everywhere get something for their capital.

                AI TRADING ROBOT

                AI trading robots: the rational choice

                The Tianma robot combines several techniques into one return model. Its speed, strategy selection, market judgement, position control and risk management all sit far ahead of a manual desk.

                Neural networksLarge-scale statisticsProprietary algorithmsCapital-flow analysis

                Technical

                • Speed: market data processed far faster than any human desk
                • Selection: the best strategy is chosen automatically for each regime
                • Judgement: key opportunities and risk points identified
                • Risk: stops set automatically, single-trade loss capped

                For the investor

                • Time back: no need to watch screens all day
                • Discipline: signals followed without emotion
                • Capital management: allocation adjusted to conditions
                • Sleep: the system monitors while you work

                Knowing when to buy and sell is not enough. You also need to know how much — position sizing and risk control are half the job.

                MISSION

                The road is long — we keep searching

                MISSION

                Something for every investor

                Apply AI to quantitative trading so that professional capability is no longer reserved for a few.

                  VISION

                  A leader in quantitative trading

                  Set the direction for intelligent investing and turn research into durable infrastructure.

                    VALUE

                    Lower the barrier with technology

                    Let AI create durable returns for investors instead of more noise.