Omitted Variable Bias, Unobserved Heterogeneity, and Endogeneity
Omitted variable bias occurs when a regression model omits one or more relevant variables, as a result, the variable gets reflected in the error term which leads to biased and counterintuitive estimates. It happens when a predictor is excluded and it gets reflected in the error term which leads to biased and counterintuitive estimates. It can be detected by checking the correlation between predictors.
Omitted Variable bias violates the linear regression assumption that “ Explanatory Variables must be exogenous “. That is error term should not contain any information about explanatory variables.
Omitted variables bias occurs as many individual characteristics are not observed.
For example, if I have a dataset to predict the Revenue of an Airline with predictor, Airline Name, Year, number of passengers traveling, Fuel price, and Load factor(the average capacity utilization of the fleet). But here the regression model does not consider inflight services and ground services provided by an airline to the customers which have an emotional impact on choosing the airline, management attitude, location, terminal of operation of the airline, airline policies, etc. These unconsidered factors are called unobserved heterogeneity.
When an explanatory variable is correlated with the error term it is referred to as endogeneity or the explanatory variable is endogenous.
