GLS回归的意义？

是 否 +2 论坛币

k人 参与回答

经管之家送您一份

应届毕业生专属福利!

求职就业群

赵安豆老师微信：zhaoandou666

经管之家联合CDA

送您一个全额奖学金名额~ !

立即领取

感谢您参与论坛问题回答

经管之家送您两个论坛币！

+2 论坛币

                                                                                                                                                [size=8.000000pt]Hence, the second step is to test for taste-driven gender discrimination. In this case, we exploit the longitudinal dimensionof our dataset using panel data estimators to control for unobserved individual effects which could systematically differ acrossentrepreneurs. However, as gender is time-invariant, in order totest for discrimination, we can only use GLS estimator assumingindividual effects to be random. GLS regression results reportedin [size=8.000000pt]Table 4 [size=8.000000pt]show that in the models of interest rate, collateral aswell as incidence of overdrawing, the coefficients on the gendervariable are very similar in sign, magnitude and statistical signifi-cance to those obtained in pooled regressions (see column 1 in[size=8.000000pt]Table 3[size=8.000000pt], and columns 3 and 6 in [size=8.000000pt]Table 4 [size=8.000000pt]where for the sake of com-parison we report results for linear probability models).  各位大神，这是我在论文里看到的一段话，按照我粗浅的理解就是，因为性别是不随时间变化的，所以我们要用GLS来使性别随机化。 可我在陈强的计量经济学上看到，GLS不是用来处理异方差和自相关的嘛？和这里的处理让性别随机化的处理有什么关系呢？？   求各位大神赐教！
                               
                       
               

扫码加我 拉你入群

请注明：姓名-公司-职位

以便审核进群资格，未注明则拒绝

分享0 收藏0 回帖