Author
Faculty member of the Department of Economics, Shiraz University
Abstract
The distributed lag effect of a unit change in one of the explanatory variables on the dependent variable is one of the major shortcomings of the standard linear regression model. Such a model, that specifies a causal relationship between a variable and its determinants, states that a unit change in one of the explanatory variables can result in a change in the dependent variable only during the period specified by the model. In practice, however, changes in, for example, the firm's advertisement may affect its sales over various periods.
This paper aimes to develop an approach to estimate the distributed lag effects of a temporary (once -and - for - all) and a permanent (continuous) change in the explanatory variables by using information given by a cointegration model.
)