In this study, we conducted a multilevel analysis of factors affecting customer satisfaction in the global hotel industry. The survey data collected from TripAdvisor.com included customer reviews relating to 13,410 hotels located in 80 major global urban tourism destinations. We examined multiple relevant factors at each of the following five levels of analysis: (a) service encounter, (b) visitor, (c) visitor’s nationality, (d) hotel, and (e) destination. The results show that hotel attributes and the personal characteristics of visitors most powerfully influence customer satisfaction. However, the purpose of the trip, the characteristics of the destination, and the visitor’s nationality are also found to play an important role in hotel evaluation. By integrating multiple levels of analysis into a single statistical model, multilevel modeling framework enables researchers and professionals to see the “big picture” of factors affecting customer satisfaction in the contemporary hotel industry.

The aim of this paper is to present different models for predicting the possibility of opening bankruptcy proceedings in companies in Serbia, as well as to research which models are most suitable for companies in the agricultural sector. In this paper, we have used and displayed three models: the Altman 's Z-score model, Kralicek's DF model and Quick test. Many authors have dealt with this issue, but most of them have developed models for developed markets, which are different from market of Serbia. Striving towards improving the analysis and prediction of bankruptcy has led to comparison of the reference value, in order to obtain concrete models for the evaluation of difficulty in the functioning of the company. In this connection, on a sample of five companies operating on the territory of the Republic of Serbia, we have applied three models that used standard financial indicators to show the financial condition and stability of the company. Results suggest that Kralicek's DF model indicates better financial state of the company than Altman 's Z-score model, considering the characteristics of the market in which the model is formed.

The goal of this study is to overcome the identified methodological limitations of prior studies aimed at predicting the type of auditor opinion and draw definite conclusions on the relative predictive performance of different predictive methods for this particular task. Predictive performance of twelve candidate models from the realms of statistics and machine learning is assessed separately for the two common real-life scenarios: a) when prior information on the client (i.e. types of audit opinion received in the past) is available and can be used in prediction, and b) when such information is not available (e.g. new companies). The results show that, in the first scenario, several methods from both realms achieve comparable predictive performance of around 0.89, as measured by the Area under the curve (AUC). In the second scenario, however, machine learning algorithms, particularly tree-based ones, such as random forest, perform significantly better, achieving the AUC of up to 0.79. Finally, we develop and assess the predictive performance of two hybrid models aimed at combining the strong points of both statistical (i.e. interpretability of results) and machine learning (i.e. handling a large number of predictors and improved accuracy) approaches. The complete procedure is demonstrated in a reproducible manner, using the largest empirical data set ever used in this stream of research, comprising 13,561 pairs of annual financial statements and the corresponding audit reports. The procedures described in this study allow audit and finance professionals around the globe to develop and test predictive models that will aid their procedures of audit planning and risk assessment.