|
The following 20 observations are drawn from a censored normal distribution:
The applicable model is y*i = μ + εi, y + i = y*i if μ + εi > 0, 0 otherwise, εi ~ N [0, σ2]. Exercises 1 through 4 in this section are based on the preceding information. The OLS estimator of μ in the context of this To bit model is simply the sample mean. Compute the mean of all 20 observations. Would you expect this estimator to over or underestimate μ if we consider only the nonzero observations, then the truncated regression model applies. The sample mean of the non limit observations is the least squares estimator in this context. Compute it and then comment on whether this sample mean should be an overestimate or an underestimate of the true mean. |
| New search. (Also 1294 free access solutions) |