Consider the full sample (solid black/horizontal line) and the rolling window (solid blue line) estimates of the Okun's law by regressing the change in the seasonally adjusted unemployment rate on real GDP growth. The dashed blue lines are the 95% pointwise confidence interval for the Okun's coeficient (slope coeficient) using non-HAC standard errors, while the dashed red lines are the 95% pointwise confidence interval using HAC standard errors. Regression of change in unemployment on output growth 0.0 -0.2 Coefficient -0.3 -0.4 -0.5 -0.6 1970 1980 1990 2000 2010 Index 1. What do HAC-based confidence intervals suggest about the uncertainty in the estimates of the slope coefficient? 2. What do the rolling window estimates of the Okun's coefficient suggest about the stability of the relationship? Your answer should discuss both statistical and economic interpretation of the results. 3. Provide one reason why a recursive estimation might have been preferred to the rolling window estimation