In the realm of financial analysis, ordinary least squares (OLS) estimation is a widely used technique to understand the behavior of financial variables. A financial analyst must carefully select appropriate regressors to include in the OLS model to ensure accurate results. Omitted variable bias occurs in this context. Can you explain the specific conditions under which omitted variable bias arises?