Mike G. Tsionas

Bayesian Approach for the Measurement of Tourism Performance A Case of Stochastic Frontier Models

Despite its rapid growth across several social science disciplines, the use of the Bayesian approach to measure tourism performance has yet to gain strong attention in tourism research. This article reviews the foundation of the Bayesian approach and discusses its benefits and the flexibility it provides in the estimation of highly complicated performance models. With the lack of tourism studies focusing on the Bayesian approach, we take first a general approach and provide a description of the Bayesian approach, illustrating its advantages, and its key differences from the frequentist approach. We then discuss its specific benefits in the measurement of tourism performance within the context of stochastic frontier (SF) models. We introduce several advanced versions of SF where the use of the Bayesian approach becomes necessary. We also provide simulation evidence about the advantages of the Bayesian approach and discuss how it can be used to estimate various SF models.

Endogenous bank risk and efficiency

We develop a framework to incorporate bank risk, as measured from the variance of profits or returns, within a model of frontier efficiency. Our framework follows the premise that risk is endogenously related to efficiency. We estimate our model using panel data for U.S. banks and Bayesian techniques. We show that excluding risk from the efficiency model significantly biases the efficiency estimates and the ranking of banks according to their competitive advantage. We also demonstrate that there is a negative risk-efficiency nexus with causality running both ways, while our estimates of risk are fully consistent with the developments in the banking industry over the period 1976–2014.

Zero-inefficiency stochastic frontier models with varying mixing proportion: A semiparametric approach

In this paper, we propose a semiparametric version of the zero-inefficiency stochastic frontier model of Kumbhakar, Parmeter, and Tsionas (2013) by allowing for the proportion of firms that are fully efficient to depend on a set of covariates via unknown smooth function. We propose a (iterative) backfitting local maximum likelihood estimation procedure that achieves the optimal convergence rates of both frontier parameters and the nonparametric function of the probability of being efficient. We derive the asymptotic bias and variance of the proposed estimator and establish its asymptotic normality. In addition, we discuss how to test for parametric specification of the proportion of firms that are fully efficient as well as how to test for the presence of fully inefficient firms, based on the sieve likelihood ratio statistics. The finite sample behaviors of the proposed estimation procedure and tests are examined using Monte Carlo simulations. An empirical application is further presented to demonstrate the usefulness of the proposed methodology.

Unobserved Heterogeneity in Hospitality and Tourism Research

Despite the growing complexity of structural equation model (SEM) applications in tourism, it is surprising that most applications have estimated these models without accounting for unobserved heterogeneity. In this article, we aim to discuss the concept of unobserved heterogeneity in more detail, highlighting its serious threats to the validity and reliability of SEMs. We describe a Bayesian finite mixture modeling framework for estimating SEMs while accounting for unobserved heterogeneity. We provide a comprehensive description of this model, and provide guidance on its estimation using the WinBUGS software. We illustrate the importance of unobserved heterogeneity and the finite mixture modeling framework using a didactic application on brand equity where heterogeneity is likely to play an important role because of the differences in how consumers perceive the different dimensions of brand equity. We compare between various models and illustrate the differences between the standard and heterogeneous SEM and discuss the implications for research and practice.

Parameters measuring bank risk and their estimation

The paper develops estimation of three parameters of banking risk based on an explicit model of expected utility maximization by financial institutions subject to the classical technology restrictions of neoclassical production theory. The parameters are risk aversion, prudence or downside risk aversion and generalized risk resulting from a factor model of loan prices. The model can be estimated using standard econometric techniques, like GMM for dynamic panel data and latent factor analysis for the estimation of covariance matrices. An explicit functional form for the utility function is not needed and we show how measures of risk aversion and prudence (downside risk aversion) can be derived and estimated from the model. The model is estimated using data for Eurozone countries and we focus particularly on (i) the use of the modeling approach as a device close to an “early warning mechanism”, (ii) the bank- and country-specific estimates of risk aversion and prudence (downside risk aversion), and (iii) the derivation of a generalized measure of risk that relies on loan-price uncertainty. Moreover, the model provides estimates of loan price distortions and thus, allocative efficiency.

Modeling and Forecasting Regional Tourism Demand Using the Bayesian Global Vector Autoregressive (BGVAR) Model

Increasing levels of global and regional integration have led to tourist flows between countries becoming closely linked. These links should be considered when modeling and forecasting international tourism demand within a region. This study introduces a comprehensive and accurate systematic approach to tourism demand analysis, based on a Bayesian global vector autoregressive (BGVAR) model. An empirical study of international tourist flows in nine countries in Southeast Asia demonstrates the ability of the BGVAR model to capture the spillover effects of international tourism demand in this region. The study provides clear evidence that the BGVAR model consistently outperforms three other alternative VAR model versions throughout one- to four-quarters-ahead forecasting horizons. The potential of the BGVAR model in future applications is demonstrated by its superiority in both modeling and forecasting tourism demand.

Microfoundations for stochastic frontiers

The purpose of the paper is to propose microfoundations for stochastic frontier models. Previous work shows that a simple Bayesian learning model supports gamma distributions for technical inefficiency in stochastic frontier models. The conclusion depends on how the problem is formulated and what assumptions are made about the sampling process and the prior. After the new formulation of the problem it turns out that the distribution of the one-sided error component does not belong to a known family. Moreover, we find that without specifying a utility function or even the cost inefficiency function, the relative effectiveness of managerial input can be determined using only cost data and estimates of the returns to scale. The point of this construction is that features of the inefficiency function u(z) can be recovered from the data, based on the solid microfoundation of expected utility of profit maximization but the model does not make a prediction about the distribution.

Notes on technical efficiency estimation with multiple inputs and outputs

Collier, Johnson and Ruggiero (2011) deal with the problem of estimating technical efficiency using regression analysis that allows multiple inputs and outputs. This revives an old problem in the analysis of production. In this note we provide an alternative maximum likelihood estimator that addresses the concerns. A Monte Carlo experiment shows that the technique works well in practice. A test for homotheticity, a critical assumption in Collier, Johnson and Ruggiero (2011) is constructed and its behavior is examined using Monte Carlo simulation and an empirical application to European banking.

Estimating Technical Efficiency and Production Risk under Contract Farming: A Bayesian Estimation and Stochastic Dominance Methodology

We investigate production risk, technical efficiency and risk attitudes amongst contract and independent farmers. We use a Bayesian parametric approach and stochastic dominance quantile regression methods to compare technical efficiency and risk attitude of smallholders in Nepal. Using farm‐level data, we find that contract farmers appear to show lower inefficiency and lower production risk. Additionally, contract and independent farmers can increase output by reducing the scale of operation. Regardless of the commodity produced and farming arrangement (contract or independent production), we find that labour, land and other inputs are risk‐augmenting, while the role of capital is mixed. We find a second order stochastic dominance (SSD) for lentils, and first order stochastic dominance (FSD) for tomatoes, ginger and HYV paddy seed commodities. Finally, contract farmers are more risk averse than independent farmers, regardless of the commodity produced.

A novel model of costly technical efficiency

This paper presents a novel model of measuring technical inefficiency based on the notion that higher efficiency requires a certain cost. First, we apply the “rational inefficiency hypothesis” of Bogetoft and Hougaard (2003) but we fail to find that it rationalizes our data set of large U.S banks with multiple inputs and outputs. In consequence, we adopt a novel model of profit maximization which explicitly incorporates the cost of technical inefficiency. The cost of inefficiency is treated as unknown and is parametrized as a function of inputs, outputs and decision-making-unit specific fixed effects. More importantly, by showing the model to be equivalent to one in which inefficiency is an arbitrary function of inputs, outputs and the inefficiency cost, we are able to determine optimal directions in the input-output space that would reduce inefficiency. Bayesian techniques organized around Markov Chain Monte Carlo are used to perform the computations and provide statistical inferences as well as useful policy measures to reduce inefficiencies in the U.S banking sector through an examination of different realistic scenarios.