Comparative Accuracy and Explanatory Performance of Valuation Models : Evidence from India

The purpose of this paper is to empirically examine the comparative accuracy and explanatory performance of discounted cash flow (DCF), residual income model (RIM), equity dividend (ED), P/E multiple (PE_M) and P/B multiple (PB_M) valuation models for the Indian banking sector and come up with a composite valuation model (CV) to see whether combining value estimates increase the valuation accuracy. To achieve the objective of the study we determined the intrinsic values using all the six models. Further, we compared the models based on prediction errors and the explanatory performance of market value on value estimates. The study uses panel regression to forecast estimates of earnings and measure explanatory performance. The study uses annual data points starting from March 31,2002 to March 31, 2012.The comparative framework shows that the most appropriate method for value estimate is provided by RIM and ED models and therefore has higher ability to account for long term market expectations for the banking sector whereas composite value estimates stay in between DCF, RIM and ED models (best three) and prescribes a middle path. Hence, combining make sense because in volatile emerging economies it is always good to follow a midway path to avoid extreme values. This paper provides academicians and practitioners with a snapshot of the applicability of DCF, RIM, ED, PE_M and PB_M valuation models for Indian banking industry and also shows how a composite value (CV) estimate can improve valuation accuracy