This research targets truncated normal dependent variable with TSCS (time-series-cross-section) data property. It is one of the studies related to my previous work, "Fixing Boundary Violations: Applying Constrained Optimization to the Truncated Regression Model". While the main issue is to solve boundary violations, TSCS data property further complicates the way how the technique of constrained optimization can be applied. In a nutshell, the current method, panel regression, for analyzing panel data with a truncated normal dependent variable suffers serious methodological problems in three aspects: boundary violations, parameter estimation, and model specification. My empirical finding discovers that the current method is sensitive to different centering methods and tends to generate false significance results. The purpose of this research is to tackle this problem by proposing a revised panel regression method. A comparative study is also carried out to provide useful methodological suggestions.