一同學
2018-07-29 11:35Assume that in a particular multiple regression model, it is determined that the error terms are uncorrelated with each other. Which of the following statements is most accurate? A. Serial correlation may be present in this multiple regression model, and can be confirmed only through a Durbin -Watson test. B. Multicollinearity exists in this multiple regression model, and can be corrected through the addition of a correlated variable. C. This model is in accordance with the basic assumptions of multiple regression analysis because the errors are not serially correlated. D. Unconditional heteroskedasticity present in this model should not pose a problem, but can be corrected by using robust standard errors. 老師您好！請問A選項中的serial correlation指的到底是Yt越Yt-i之間相關，還是et與et-i之間相關？在課程中老師說指的是后者，這道題的解析里，老師說是前者。能否明確一下哪一個是對的？ 還有D選項中，Conditional和Unconditional分別會對模型產(chǎn)生什么影響？ 謝謝老師！
所屬：FRM Part I 視頻位置 相關試題
來源： 視頻位置 相關試題
1個回答
Crystal助教
2018-07-30 11:20
該回答已被題主采納
同學你好，兩個老師說的都是正確的，因為在時間序列中y表示的其實就是殘差項，這個講義中我們是有寫的。 條件異方差和非條件異方差都是不會影響到斜率的取值的，影響的僅僅是標準誤，進而會影響到t統(tǒng)計量的數(shù)值。
0
0
0
0
