Suppose that we have a cross-sectional data set collected at 2020. In the data set, the dependent variable is education level of an individual. The sampled individuals are aged from 25 to 30.

In this setting, I am trying to predict the final education level of individuals who are not included in the sample.

In this situation, I suddenly thought "what if the dependent variable changes overtime?" That is, my problem is that we cannot know the "real" final education level when we fit a prediction model.

More specifically, think about the possibility that an individual who was high-school graduates at 2020, the data collection year, can become a college graduate in the next year. If we use just the 2020 data for model fitting and many people become higher educated between age 25 and 30, we cannot accurately predict other individuals final education level using our model fitting results.

In this case, how can I mitigate or overcome this problem??



Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge that you have read and understand our privacy policy and code of conduct.

Browse other questions tagged or ask your own question.