This study examines the information efficiency of eight stock valuation models, including a novel model called NAPV, to accurately estimate stock prices by reflecting all relevant information. An experiment was conducted with 65 traders, 20 evaluators, and a virtual company. The evaluators provided an informationally efficient benchmark. The stock price growth rates generated by each model were compared to this benchmark. The results indicate that NAPV was the most efficient model, accounting for 89% of the variation in the benchmark. Meanwhile, the Constant-Growth DDM, Adjusted Net Asset, and Constant-Growth RIM models explained approximately 30% of the variation. Other models showed no significant relationship. Income-based models were less efficient than other models, some generating no value at times. The two-stage model’s sensitivity to estimated terminal value limits its efficiency. The findings suggest that NAPV is a promising tool for valuation, as it can decompose the stock value into key components of value. NAPV’s full potential requires some regulatory measures and an updated stock pricing mechanism. Implications include shareholders carefully choosing a model, regulators enabling NAPV’s potential, and developing new models addressing limitations. Future research could explore other influences on models’ efficiency, study implementing regulatory measures’ impact, and develop new models.

Stock valuation models, Information efficiency, NAPV model, Growth premium, Evaluators