The paper aims at the empirical evaluation of the impact of bank size (as measured by median total assets) on the value relevance of two key accounting variables, i.e. book values of equity and net earnings, in terms of their joint explanatory power in the regression model and the relative responsiveness of bank market values to the changes in those variables. The research is based on the multiple linear regression analysis after controlling for the presence of fixed and random effects. The examined sample covers all domestically-based commercial banks listed on the Warsaw Stock Exchange over the period 1998–2017. The final pooled sample comprises 18 banks and 271 bank-year observations. The findings of the study suggest that the equity investors perceive the joint informational content of book values and earnings of larger banks as more value relevant in comparison to the accounting numbers reported by their smaller peers. The responsiveness of banks’ market values to the changes in each of the explanatory variables seems, however, to be affected by their size in a different way. As expected, book values of equity have turned out to be significantly more informative for smaller banks, whereas the evidence regarding the impact of size on the responsiveness of bank market values to the changes in net earnings is ambiguous. Although larger banks appear to exhibit a higher sensitivity of stock prices to variations in net earnings per share than their smaller peers, the difference between the examined subsamples is not statistically significant.

Inaccurate estimation in highway projects represents a major problem facing planners and estimators, especially when data and information about the projects are not available, and therefore the need to use modern technologies that addresses the problem of inaccuracy of estimation arises. The current methods and techniques used to estimate earned value indexes in Iraq are weak and inefficient. In addition, there is a need to adopt new and advanced technologies to estimate earned value indexes that are fast, accurate and flexible to use. The main objective of this research is to use an advanced method known as artificial neural networks to estimate the TSPI of highway buildings. The application of artificial neural networks as a new digital technology in the construction industrial in Republic of Iraq is absolutely necessary to ensure successful project management. One model built to predict the TCSPI of highway projects. In this current study, artificial neural network model were used to model the process of estimating earned value indexes, and several cases related to the construction of artificial neural networks have been studied, including network architecture and internal factors and the extent of their impact on the performance of artificial neural network models. Easy equation was developed to calculate that TSPI. It was found that these networks have the ability to predict the TSPI of highway projects with a very outstanding saucepan of reliability (97.00%), and the accounting coefficients (R) (95.43%).