上财会计与财务研究方法论课件(李增泉教授).ppt

* 属性数据的分析 变量的类型 名义变量和有序变量：属性（离散）变量 区间变量和比例变量：定量（连续）变量 列连表分析 属性变量的无关联性分析（X2检验） 有序变量的关联性分析 SAS：PROC FREQ * 回归分析 估计量的性质 无偏性：E（b^）=b 有效性：min(var(b^)) 最小平均偏差平方 一致性：b^依概率收敛于b OLS 无偏、有效 * 回归分析（2） 假设检验 回归方程的显著性：F 回归系数的显著性：t 模型的拟合度：R2/Adj-R2(T1) * 回归分析（3） 回归诊断 基本假定 Y与Xi线性相关， Xi之间不存在线性关系 ei相互独立 E(ei)=0, var(ei)=q2 Ei分布为N（0，q2） * 回归分析（4） 回归诊断 共线性 异方差 自相关 极值的处理 1%剔除 WINSORIZE 分析残差 * 回归分析（5） 几个特殊的问题 比较两个模型的R2 VUONG TEST 比较几个回归系数 F TEST * 回归分析（6） 典型的SAS程序 Proc reg NOINT; Model Y=A B C/VIF SPEC ACOV DW TEST A=B C=0.5; OUTPUT OUT=data COOKD=ckd Dffits=dft rstudent=rst; Run; * On the use of instrumental variables in accounting research David F.Larcker TjommeO.Rusticus Journal of Accounting and Economics 49 (2010) 186–205 Instrumental variable (IV) methods are commonly used in accounting research (e.g., earnings management, corporate governance, executive compensation, and disclosure research) when the regressor variables are endogenous. when the outcome and explanatory variables are simultaneously determined (i.e., simultaneous-equation bias) when a variable that affects both the outcome and explanatory variables is not included in the regression model (i.e., correlated omitted variable bias) This solution to endogeneity is appropriate if the researcher can find instrumental variables that are correlated with the endogenous regressor but uncorrelated with the error in the structural equation Two potential problems can arise with the instrumental variable estimator First,if the instrument is not actually exogenous the IV estimate moves further away from the true value of b，This problem is exacerbated when corr(x, z) is small(i.e., when the instruments’ first-stage explanatory power is small). Second, in finite samples, when corr(x, z) is close to zero, the estimates are very unreliable and the standard test statistics are misspecified Assuming that IV approaches are used, there are several fundamental requirements that must be met for any IV application in accounting research Fir