x*******u
发帖数: 500 | 1 linear regression model 对y 有normal的要求, 请问它对x有没有要求? 如果x r
ight skew, 是不是一定要tranform 或者改成categorical variable? 请大牛指教
. 谢谢. |
m****e
发帖数: 255 | 2 General you want X to be roughly normal. But in reality as the residuals are
normal, you are fine. Otherwise, transform, categorical or nonparametric
regression. |
D******n
发帖数: 2836 | 3 no
【在 x*******u 的大作中提到】
: linear regression model 对y 有normal的要求, 请问它对x有没有要求? 如果x r
: ight skew, 是不是一定要tranform 或者改成categorical variable? 请大牛指教
: . 谢谢.
|
q**j
发帖数: 10612 | 4 small sample needs the noise to be normal. but most of the time we care abou
t large sample where nomality is not need for either noise or x. you only ne
ed: x(i) is bounded, 1/n*sum(x(i)x(i)') converges and the noise are uncorrel
ated with x and independent themselves.
【在 x*******u 的大作中提到】
: linear regression model 对y 有normal的要求, 请问它对x有没有要求? 如果x r
: ight skew, 是不是一定要tranform 或者改成categorical variable? 请大牛指教
: . 谢谢.
|
t*******t
发帖数: 633 | 5 OLS中Xis not random!!!it's a variable but not random variable |
i******n
发帖数: 839 | 6 aglee.
【在 t*******t 的大作中提到】
: OLS中Xis not random!!!it's a variable but not random variable
|
n*****1
发帖数: 172 | 7 I guess you can treat X as random variables in OLS. In that case, there is
not much to talk about the finite sample properties of the betas. For
asymptotic properties, you need additional (stronger) assumptions about X
and e to derive consistency. And I don't think you need to transform the
data ONLY because it is skewed. But that also depends on what your
underlying economic model is. |
m*****s
发帖数: 1381 | 8 对error term的要求只是 0 mean。 Normality 不会影响参数统计的结果，不过会降低
efficiency。
Normality成立的话，ML和LS两种方法得出的结果是一致的。
【在 x*******u 的大作中提到】
: linear regression model 对y 有normal的要求, 请问它对x有没有要求? 如果x r
: ight skew, 是不是一定要tranform 或者改成categorical variable? 请大牛指教
: . 谢谢.
|
d******e
发帖数: 7844 | 9 Normality成立的话,ML和LS对beta的估计是一致的，但是ML的Sigma^2是biased.
【在 m*****s 的大作中提到】
: 对error term的要求只是 0 mean。 Normality 不会影响参数统计的结果，不过会降低
: efficiency。
: Normality成立的话，ML和LS两种方法得出的结果是一致的。
|
o****o
发帖数: 8077 | 10 in standard OLS, you model P(Y|X), not P(Y, X)
if your X is a random variable, then you are doing measurements error model,
and in that case, generally X is required to have a normal distribution
through its own random disturbance
【在 x*******u 的大作中提到】
: linear regression model 对y 有normal的要求, 请问它对x有没有要求? 如果x r
: ight skew, 是不是一定要tranform 或者改成categorical variable? 请大牛指教
: . 谢谢.
|
