0

According to https://en.m.wikipedia.org/wiki/Ordinary_least_squares, the OLS estimator gives each individual's dependent variable a weight, and the weights depend on regressors like below:

As was mentioned before, the estimator $ {\hat {\beta }}$ is linear in $y$, meaning that it represents a linear combination of the dependent variables $y_i$. The weights in this linear combination are functions of the regressors $X$, and generally are unequal.

Here, I am not sure how the OLS estimator gives each $y_i$ the weights that consist of $X$.

Could you show me the precise expression of the linear combination mentioned above?

QWEQWE
  • 471
  • 1
  • 9
  • 1
    I am confident you are familiar with the standard formula $\hat \beta=(X^\prime X)^{-}X^\prime y.$ That's the product of a matrix on the left hand side by the vector $y:$ by definition, *it is a linear combination of $y.$* Thus, what you are asking for has appeared here on CV very many times. In the duplicate this fact is explicitly pointed out. I found a second duplicate that uses a geometric argument and have added that to the list. – whuber Nov 05 '21 at 22:48

0 Answers0