Skip to main content
4 votes
Accepted

Finding the distribution in OLS

The distribution will be $t_{n-2}$. Below, I assume the sample $\{x_i\}_{i=1}^n$ is treated as fixed. That is, I do all computations as if we were conditioning on $\{x_i\}_{i=1}^n$. I relax this ...
Joseph Basford's user avatar
2 votes

In a regression of Yon X, the value of X is fixed at 5. Find the regression equation

If there is no constant in the regression, then $$ \widehat{\beta} = \frac{\sum_i x_iy_i}{\sum_i x_i^2} = 5\frac{\sum_i y_i}{5^2N} = \frac{\bar{y}}{5}.$$ It is not a slope, however, as $x_i$ is ...
Bertrand's user avatar
  • 3,371
2 votes

In a regression of Yon X, the value of X is fixed at 5. Find the regression equation

If $X$ is constant it is impossible to find the regression equation because regression coefficients $\beta_i$ are given by: $$\beta_i = \frac{cov(Y, X_i)}{var(X_i)}$$ if you fix $X$ to be always 5 the ...
1muflon1's user avatar
  • 56.7k
2 votes
Accepted

Difference between unconfoundedness and parallel trends

You're pretty much correct - the two assumptions are very similar. The only difference is that the Parallel Trends assumption allows for differences between the treatment group and the control group (...
matthewoulton's user avatar
1 vote

Random Utility Model Multiple Choice Question. Which one is correct?

I believe the correct answer is (C). Both coefficients are identifiable. The cross-sectional variation identifies the alternative-specific coefficients. It is probably easiest to see why if you assume ...
Jesper Hybel's user avatar
  • 3,386
1 vote

Two endogenous variables and two control functions

I think you could proceed by regressing $Y$ on $X$, $Z$ and $V = (V_1 + V_2)$. Note that by substituting $U$ into the regression you get: $$ Y = \alpha X + \beta Z + \gamma \delta_4 V + (\varepsilon + ...
tdm's user avatar
  • 12.2k

Only top scored, non community-wiki answers of a minimum length are eligible