Questions tagged [omitted-variable-bias]

The tag has no usage guidance.

7 questions with no upvoted or accepted answers
Filter by
Sorted by
Tagged with
2
votes
1answer
13 views

Instrument validity: does a positive and significant coefficient on Z in a regression of Y on X and Z pose a problem?

I have an initial regression of Y on X and Z. Both of my coefficients on X and Z are non-zero and strongly statistically significant. X and Z are correlated but I am told collinearity shouldn't be an ...
2
votes
1answer
121 views

How to test whether OVB by examining two regressors (X_1, X_2) using hypothesis test with null hypothesis H0: corr(X_1,X_2) = 0

Suppose you have an i.i.d. sample {(π‘Œ , 𝑋 , 𝑋 ): 𝑖 = 1, ... , 𝑛}. You want to estimate the causal effect of 𝑋1 on π‘Œ. You first run a regression π‘Œ = 𝛼0 + 𝛼1𝑋i + 𝑒i and get the following ...
2
votes
0answers
44 views

Using an IV when there is more than one omitted variable

I am trying to estimate the following model: $$y=B_0 + B_1x_1 + B_2x_2 + B_3x_3 + e$$ However, I have an omitted variable bias because $x_2$ and $x_3$ are not observed. Situation 1 If I have an (...
0
votes
0answers
42 views

Homoskedasticity

From Bruce Hansen's book, chapter 2, ex. 2.22 Suppose we have the following: $$Y=X_1'\beta_1+X_2'\beta_2+e$$ $$E[e|X_1,X_2]=0$$ $$E[e^2|X_1,X_2]=\sigma^2$$ $$E[X_2|X_1]=\Gamma X_1$$ We will assume $\...
0
votes
0answers
54 views

VAR model variable selection

I'm required to use two time series models in my exam project. I want to use a stock price of an energy company, and then explain it first using ARIMA, and then adding other variables and using VAR. ...
0
votes
0answers
13 views

Question about regression and deriving omitted variables

usually when I see derivations of ommited variable bias, I see something of the sort: from y=xb + $\eta$, and looking at the for formula for the slope estimate: $cov(x,y)$$/var(x)$ $cov(x,xb+\eta )$$/...
0
votes
0answers
49 views

Are coefficients that are zero omitted variable bias?

If a regression coefficient is essentially zero, doesn't that imply that there is (massive) omitted variable bias? That is, the change must then exist in the error term. The classic definition of OVB ...