版本 / Version: 4.0 (2025-07-07)
Language / 語言: English | 中文
An experimental economics platform built on the oTree framework, specifically designed to study the impact of different carbon reduction policies on firm production behavior.
This platform supports four experimental treatment groups, providing a comprehensive research environment for carbon emission policies:
- Control Group: Baseline experiment without carbon emission restrictions
- Carbon Tax Group: Policy experiment with carbon tax based on emission levels
- Carbon Trading Group: Carbon permit market experiment with real-time trading functionality
- MUDA Practice Group: Trading system operation training experiment
Core features include real-time trading system, intelligent matching engine, comprehensive data tracking, flexible configuration management, and modern user interface.
- Python 3.7 or higher
- oTree 5.10.0 or higher
- PostgreSQL (production) or SQLite (development)
- Download the project
git clone <repository-url>
cd Carbon-Emissions-Trading-Experiment- Create virtual environment
python -m venv venv
# Windows
venv\Scripts\activate
# macOS/Linux
source venv/bin/activate- Install dependencies
pip install -r requirements.txt- Initialize and start
otree resetdb
otree devserverAfter starting, visit http://localhost:8000 to begin the experiment.
The platform supports two operating modes, which can be switched by editing configs/experiment_config.yaml:
experiment_mode:
test_mode_enabled: true # true = test mode, false = production modeMode Comparison
| Item | Test Mode | Production Mode |
|---|---|---|
| Players per group | 2 players | 15 players |
| Number of rounds | 3 rounds | 15 rounds |
| Dominant firms | 1 firm | 3 firms |
| Trading time | 60 seconds | 120-180 seconds |
The configuration file contains complete experimental settings including firm role parameters, carbon tax settings, and trading parameters:
general:
dominant_firm:
mc_range: [1, 5] # Marginal cost coefficient range
emission_per_unit: 2 # Carbon emission per unit
max_production: 20 # Maximum production capacity
non_dominant_firm:
mc_range: [2, 7]
emission_per_unit: 1
max_production: 8
stages:
carbon_tax:
tax_random_selection:
rates: [1, 2, 3] # Carbon tax rate options
carbon_trading:
trading_time: 120 # Trading time (seconds)
carbon_allowance_per_player: 10 # Initial carbon allowanceExperimental Flow: Introduction → Production Decision → Results Display
Features: No carbon emission restrictions, establishing baseline data for pure market mechanisms
Experimental Flow: Introduction → Production Decision → Results Display
Features: Carbon tax levied based on emission levels, with tax calculation formula: Carbon Tax = Emission Level × Tax Rate
Experimental Flow: Introduction → Carbon Permit Trading → Production Decision → Results Display
Features: Participants must first engage in carbon permit trading, production is limited by carbon permit holdings, using real-time matching trading mechanism
Experimental Flow: Introduction → Trading Practice → Results Display
Features: Pure trading operation practice without production decisions, used for familiarizing with the trading interface
Total cost uses an increasing marginal cost structure:
Total Cost = Σ(Marginal Cost Coefficient × i + Random Disturbance) for i = 1 to Production Quantity
- Order Types: Limit buy and sell orders
- Matching Rules: Price priority, time priority
- Trading Restrictions: Sell order quantity cannot exceed holdings
- Real-time Updates: Using WebSocket technology for real-time market status synchronization
- Production Decisions: Output, cost, revenue, profit data
- Trading Behavior: Complete records of orders, transactions, cancellations
- Market Dynamics: Price trends, trading volume, market depth changes
Carbon-Emissions-Trading-Experiment/
├── configs/ # Experimental configuration files
│ ├── experiment_config.yaml
│ └── config.py
├── utils/ # Shared utility modules (v3.1 optimized)
│ ├── shared_utils.py # Core utility functions
│ ├── trading_utils.py # Trading-specific tools (NEW in v3.1)
│ └── database_cleaner.py # Database management
├── Stage_Control/ # Control group experiment (v3.1 refactored)
├── Stage_CarbonTax/ # Carbon tax group experiment (v3.1 refactored)
├── Stage_MUDA/ # Practice group experiment (v3.1 refactored)
├── Stage_CarbonTrading/ # Carbon trading group experiment (v3.1 refactored)
├── docs/ # Related documentation
└── requirements.txt # Dependency list
The platform automatically collects experimental data and supports export in multiple formats:
- CSV Format: Suitable for statistical software analysis
- JSON Format: Suitable for programmatic processing
- Excel Format: Suitable for manual inspection and preliminary analysis
All data includes integrity checking mechanisms to ensure analysis-ready data quality.
OTREE_ADMIN_PASSWORD=your_password
OTREE_SECRET_KEY=your_secret_key
DATABASE_URL=postgresql://user:pass@host:port/dbnameFROM python:3.9
WORKDIR /app
COPY . .
RUN pip install -r requirements.txt
CMD ["otree", "prodserver", "0.0.0.0:8000"]- System Functions and Operating Logic
- Development Work Log
- Data Codebook
- Database Cleaner Tool Instructions
For technical issues or research collaboration inquiries, please submit through GitHub Issues.
基於 oTree 框架開發的經濟學實驗平台，專門用於研究不同碳減排政策對廠商生產行為的影響。
本平台支援四種實驗處理組別，提供完整的碳排放政策研究環境：
- 對照組：無碳排放限制的基準實驗
- 碳稅組：基於碳排放量徵收稅金的政策實驗
- 碳交易組：具備即時交易功能的碳權市場實驗
- MUDA 練習組：交易系統操作訓練實驗
核心功能包括即時交易系統、智慧撮合引擎、完整數據追蹤、靈活配置管理和現代化使用者介面。
- Python 3.7 或更高版本
- oTree 5.10.0 或更高版本
- PostgreSQL（生產環境）或 SQLite（開發環境）
- 下載專案
git clone <repository-url>
cd Carbon-Emissions-Trading-Experiment- 建立虛擬環境
python -m venv venv
# Windows
venv\Scripts\activate
# macOS/Linux
source venv/bin/activate- 安裝相依套件
pip install -r requirements.txt- 初始化並啟動
otree resetdb
otree devserver啟動後請造訪 http://localhost:8000 開始實驗。
平台支援兩種運行模式，可透過編輯 configs/experiment_config.yaml 進行切換：
experiment_mode:
test_mode_enabled: true # true = 測試模式, false = 正式模式模式差異對比
| 項目 | 測試模式 | 正式模式 |
|---|---|---|
| 每組人數 | 2 人 | 15 人 |
| 回合數 | 3 回合 | 15 回合 |
| 主導廠商數量 | 1 個 | 3 個 |
| 交易時間 | 60 秒 | 120-180 秒 |
配置檔案包含廠商角色參數、碳稅設定、交易參數等完整實驗設定：
general:
dominant_firm:
mc_range: [1, 5] # 邊際成本係數範圍
emission_per_unit: 2 # 每單位碳排放量
max_production: 20 # 最大生產能力
non_dominant_firm:
mc_range: [2, 7]
emission_per_unit: 1
max_production: 8
stages:
carbon_tax:
tax_random_selection:
rates: [1, 2, 3] # 碳稅率選項
carbon_trading:
trading_time: 120 # 交易時間（秒）
carbon_allowance_per_player: 10 # 初始碳權配額實驗流程：介紹 → 生產決策 → 結果顯示
特點：無碳排放限制，建立純市場機制的基準數據
實驗流程：介紹 → 生產決策 → 結果顯示
特點：根據碳排放量徵收稅金，稅額計算公式為：碳稅 = 碳排放量 × 稅率
實驗流程：介紹 → 碳權交易 → 生產決策 → 結果顯示
特點：參與者需先進行碳權交易，生產量受碳權持有量限制，採用即時撮合交易機制
實驗流程：介紹 → 交易練習 → 結果顯示
特點：純交易操作練習，不涉及生產決策，用於熟悉交易介面
總成本採用遞增邊際成本結構：
總成本 = Σ(邊際成本係數 × i + 隨機擾動) for i = 1 to 生產量
- 訂單類型：限價買單和賣單
- 撮合規則：價格優先、時間優先
- 交易限制：賣單數量不得超過持有量
- 即時更新：使用 WebSocket 技術實現即時市場狀態同步
- 生產決策：產量、成本、收益、利潤數據
- 交易行為：掛單、成交、撤單完整記錄
- 市場動態：價格走勢、成交量、市場深度變化
Carbon-Emissions-Trading-Experiment/
├── configs/ # 實驗配置檔案
│ ├── experiment_config.yaml
│ └── config.py
├── utils/ # 共用工具模組 (v3.1 優化)
│ ├── shared_utils.py # 核心工具函數
│ ├── trading_utils.py # 交易專用工具 (v3.1 新增)
│ └── database_cleaner.py # 資料庫管理
├── Stage_Control/ # 對照組實驗 (v3.1 重構)
├── Stage_CarbonTax/ # 碳稅組實驗 (v3.1 重構)
├── Stage_MUDA/ # 練習組實驗 (v3.1 重構)
├── Stage_CarbonTrading/ # 碳交易組實驗 (v3.1 重構)
├── docs/ # 相關文檔
└── requirements.txt # 相依套件清單
平台自動收集的實驗數據支援多種格式匯出：
- CSV 格式：適用於統計軟體分析
- JSON 格式：適用於程式化處理
- Excel 格式：適用於人工檢視和初步分析
所有數據均包含完整性檢查機制，確保分析就緒的數據品質。
OTREE_ADMIN_PASSWORD=your_password
OTREE_SECRET_KEY=your_secret_key
DATABASE_URL=postgresql://user:pass@host:port/dbnameFROM python:3.9
WORKDIR /app
COPY . .
RUN pip install -r requirements.txt
CMD ["otree", "prodserver", "0.0.0.0:8000"]如有技術問題或研究合作需求，請透過 GitHub Issues 提出。