A national survey estimates average household income ($\bar{y} = 68,500$) and can leverage an auxiliary variable from tax administrative data: mean household income $\bar{X}_{pop} = 72,000$.

Survey auxiliary mean: $\bar{x}_{sample} = 69,200$ Regression slope (survey income vs. tax measure): $b = 0.20$

Using the generalized regression estimator (GREG): $\bar{y}_{GREG} = \bar{y}_{sample} + b(\bar{X}_{pop} - \bar{x}_{sample})$

The calibrated GREG estimate is: