Dear all,

I am working on demand estimation for automobiles. Applying Berry's model. My dependent variable is log difference between market share and outside good. Data type is pooled cross sectional. For some observations, when i take the difference there are missing values. had anyone had same type of issue?