13 Instrumental Variables

Every result so far has rested on the assumption that the regressors are uncorrelated with the disturbance. In economics that assumption fails routinely. Price and quantity are determined together by supply and demand; income and consumption are determined together through the national accounts; education is chosen by people whose unobserved ability also affects their earnings; and most economic variables are measured with error. In each case least squares is biased and inconsistent, and the bias does not shrink as the sample grows. The method of instrumental variables restores consistency by using variables that move the regressor but are unrelated to the disturbance. This section develops the method, its efficient form, two-stage least squares, its weaknesses, and the tests that go with it.

Questions on this section

Stuck on something here? Ask below and it stays attached to this topic.