简单最小二乘回归后，预测残差时是否要加res

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

先做了简单最小二乘回归，sysuse auto, clear

regprice weight mpg turn foreign

预测残差时，老师给的步骤是 predict e, res

我自己试了一下，后面加和不加res都能预测出值但是结果不一样，有没有大神可以告知以下加res和不加res分别代表什么呢？

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

predict e  等价于 predict e,xb
预测的是线性拟合值

predict e,res
预测的是回归残差

Syntax for predict

   predict [type] newvar [if] [in] [, statistic]

    statistic            Description
    ----------------------------------------------------------------------------------------------
    Main
      xb               linear prediction; the default
      residuals          residuals
      score             score; equivalent to residuals
      rstandard          standardized residuals
      rstudent           Studentized (jackknifed) residuals
      cooksd             Cook's distance
      leverage | hat       leverage (diagonal elements of hat matrix)
      pr(a,b)            Pr(y | a < y < b)
      e(a,b)             E(y | a < y < b)
      ystar(a,b)         E(y*), y* = max(a,min(y,b))
    * dfbeta(varname)    DFBETA for varname
      stdp               standard error of the linear prediction
      stdf               standard error of the forecast
      stdr               standard error of the residual
    * covratio           COVRATIO
    * dfits              DFITS
    * welsch             Welsch distance
    ----------------------------------------------------------------------------------------------
    Unstarred statistics are available both in and out of sample; type predict ... if e(sample)
      ... if wanted only for the estimation sample.  Starred statistics are calculated only for
      the estimation sample, even when if e(sample) is not specified.
    rstandard, rstudent, cooksd, leverage, dfbeta(), stdf, stdr, covratio, dfits, and welsch are
      not available if any vce() other than vce(ols) was specified with regress.
    xb, residuals, score, and stdp are the only options allowed with svy estimation results.

    where a and b may be numbers or variables; a missing (a > .) means minus infinity, and b
      missing (b > .) means plus infinity; see missing.