| Summary: | It is not surprising that a number of economists have tried to estimate the import demand function, given its practical and theoretical importance for the field. In the article the authors substantially repeated earlier procedures, in an effort to find corroboration for previous estimates. However, they made these changes, their data period for the study was 1953-1971, quarterly data was used throughout, lagged variables were tested to see whether or not they added significantly to the predictive capacity of demand functions, a time variable was added, and alternative independent variables were tested for significance. The researchers utilized a least squares technique because it permitted the replication of prior experiments and because the case against its use did not seem strong enough to void its appearance in the relevant literature. Two identical sets of regressions were run in each sub-period, one incorporating real, and the other nominal independent variables. It seems clear that there has been a significant shift in the import demand function from the period of the fifties to the sixties. Much of this can be explained by a shift in the composition of imports.
|
|---|