aigroup-econ-mcp 1.4.3__py3-none-any.whl → 2.0.1__py3-none-any.whl

econometrics/tests/causal_inference_tests/test_instrumental_variables.py

@@ -0,0 +1,44 @@
+"""
+工具变量法测试
+"""
+
+import numpy as np
+import unittest
+from econometrics.causal_inference.causal_identification_strategy import instrumental_variables_2sls
+
+
+class TestInstrumentalVariables(unittest.TestCase):
+    
+    def test_instrumental_variables_2sls(self):
+        """测试工具变量法"""
+        # 生成模拟数据
+        np.random.seed(42)
+        n = 100
+        
+        # 工具变量
+        z = np.random.normal(0, 1, n)
+        
+        # 内生变量（与误差项相关）
+        e1 = np.random.normal(0, 1, n)
+        x = 1 + 0.5 * z + e1
+        
+        # 结果变量
+        e2 = np.random.normal(0, 1, n)
+        y = 2 + 1.5 * x + e2 + 0.3 * e1  # 包含内生性
+        
+        # 执行工具变量回归
+        result = instrumental_variables_2sls(
+            y=y.tolist(),
+            x=x.reshape(-1, 1).tolist(),
+            instruments=z.reshape(-1, 1).tolist()
+        )
+        
+        # 检查结果
+        self.assertIsNotNone(result.estimate)
+        self.assertIsNotNone(result.std_error)
+        self.assertIsNotNone(result.p_value)
+        self.assertGreater(result.n_observations, 0)
+
+
+if __name__ == "__main__":
+    unittest.main()

econometrics/tests/specific_data_modeling_tests/test_micro_discrete_limited_data.py

@@ -0,0 +1,189 @@
+"""
+微观离散与受限数据模型测试
+"""
+
+import numpy as np
+import pandas as pd
+import sys
+import os
+from scipy import stats
+
+# 添加项目根目录到路径
+sys.path.append(os.path.dirname(os.path.dirname(os.path.dirname(os.path.dirname(__file__)))))
+
+try:
+    from econometrics.specific_data_modeling.micro_discrete_limited_data import (
+        LogitModel,
+        ProbitModel,
+        TobitModel,
+        PoissonModel,
+        NegativeBinomialModel
+    )
+    HAS_MODELS = True
+except ImportError as e:
+    print(f"导入模型时出错: {e}")
+    HAS_MODELS = False
+
+
+def test_logit_model():
+    """测试Logit模型"""
+    print("测试Logit模型...")
+    
+    # 生成模拟数据
+    np.random.seed(42)
+    n = 1000
+    X = np.random.normal(0, 1, (n, 2))
+    coef_true = np.array([1.0, -0.5])
+    linear_pred = np.dot(X, coef_true)
+    prob = 1 / (1 + np.exp(-linear_pred))
+    y = np.random.binomial(1, prob)
+    
+    # 拟合模型
+    model = LogitModel()
+    model.fit(X, y)
+    
+    print(f"真实系数: {coef_true}")
+    print(f"估计系数: {model.results_.params[1:]}")  # 排除常数项
+    print(f"常数项: {model.results_.params[0]}")
+    print(f"AIC: {model.results_.aic}")
+    print(f"BIC: {model.results_.bic}")
+    print()
+
+
+def test_probit_model():
+    """测试Probit模型"""
+    print("测试Probit模型...")
+    
+    # 生成模拟数据
+    np.random.seed(42)
+    n = 1000
+    X = np.random.normal(0, 1, (n, 2))
+    coef_true = np.array([0.5, -0.3])
+    linear_pred = np.dot(X, coef_true)
+    prob = np.clip(stats.norm.cdf(linear_pred), 1e-10, 1-1e-10)
+    y = np.random.binomial(1, prob)
+    
+    # 拟合模型
+    model = ProbitModel()
+    model.fit(X, y)
+    
+    print(f"真实系数: {coef_true}")
+    print(f"估计系数: {model.results_.params[1:]}")  # 排除常数项
+    print(f"常数项: {model.results_.params[0]}")
+    print(f"AIC: {model.results_.aic}")
+    print(f"BIC: {model.results_.bic}")
+    print()
+
+
+def test_tobit_model():
+    """测试Tobit模型"""
+    print("测试Tobit模型...")
+    
+    # 生成模拟数据
+    np.random.seed(42)
+    n = 1000
+    X = np.random.normal(0, 1, (n, 2))
+    coef_true = np.array([1.0, -0.5])
+    sigma_true = 0.5
+    
+    # 生成潜在变量
+    y_latent = np.dot(X, coef_true) + np.random.normal(0, sigma_true, n)
+    
+    # 截断：低于0的值设为0
+    y = np.where(y_latent > 0, y_latent, 0)
+    
+    # 拟合模型
+    model = TobitModel(lower_bound=0)
+    model.fit(X, y)
+    
+    print(f"真实系数: {coef_true}")
+    print(f"估计系数: {model.results_.params[1:]}")  # 排除常数项
+    print(f"常数项: {model.results_.params[0]}")
+    print(f"对数似然: {model.results_.llf}")
+    print()
+
+
+def test_poisson_model():
+    """测试泊松模型"""
+    print("测试泊松模型...")
+    
+    # 生成模拟数据
+    np.random.seed(42)
+    n = 1000
+    X = np.random.normal(0, 1, (n, 2))
+    coef_true = np.array([0.5, -0.3])
+    mu = np.exp(np.dot(X, coef_true))
+    y = np.random.poisson(mu)
+    
+    # 拟合模型
+    model = PoissonModel()
+    model.fit(X, y)
+    
+    print(f"真实系数: {coef_true}")
+    print(f"估计系数: {model.results_.params[1:]}")  # 排除常数项
+    print(f"常数项: {model.results_.params[0]}")
+    print(f"对数似然: {model.results_.llf}")
+    print(f"AIC: {model.results_.aic}")
+    print(f"BIC: {model.results_.bic}")
+    
+    # 预测测试
+    y_pred = model.predict(X[:5])
+    print(f"前5个样本的预测值: {y_pred}")
+    print()
+
+
+def test_negative_binomial_model():
+    """测试负二项模型"""
+    print("测试负二项模型...")
+    
+    # 生成模拟数据
+    np.random.seed(42)
+    n = 1000
+    X = np.random.normal(0, 1, (n, 2))
+    coef_true = np.array([0.5, -0.3])
+    
+    mu = np.exp(np.dot(X, coef_true))
+    # 生成负二项分布数据
+    alpha = 0.5
+    size = 1.0 / alpha
+    prob = size / (size + mu)
+    y = np.random.negative_binomial(size, prob)
+    
+    # 拟合模型
+    model = NegativeBinomialModel()
+    model.fit(X, y)
+    
+    print(f"真实系数: {coef_true}")
+    print(f"估计系数: {model.results_.params[1:-1]}")  # 排除常数项和alpha参数
+    print(f"常数项: {model.results_.params[0]}")
+    print(f"对数似然: {model.results_.llf}")
+    print(f"AIC: {model.results_.aic}")
+    print(f"BIC: {model.results_.bic}")
+    
+    # 预测测试
+    y_pred = model.predict(X[:5])
+    print(f"前5个样本的预测值: {y_pred}")
+    print()
+
+
+if __name__ == "__main__":
+    if not HAS_MODELS:
+        print("模型不可用，请确保已安装statsmodels库")
+        exit(1)
+        
+    try:
+        import statsmodels.api as sm
+    except ImportError:
+        print("需要安装statsmodels库")
+        exit(1)
+    
+    print("微观离散与受限数据模型测试")
+    print("=" * 50)
+    
+    test_logit_model()
+    test_probit_model()
+    test_tobit_model()
+    test_poisson_model()
+    test_negative_binomial_model()
+    
+    print("所有测试完成！")