Evaluating the Behavioural Performance of Alternative Logit Models: An Application to Corporate Takeovers Research

Abstract:  Econometric models involving a discrete outcome dependent variable abound in the finance and accounting literatures. However, much of the literature to date utilises a basic or standard logit model. Capitalising on recent developments in the discrete choice literature, we examine three advanced (or non-IID) logit models, namely: nested logit, mixed logit and latent class MNL. Using an illustration from corporate takeovers research, we compare the explanatory and predictive performance of each class of advanced model relative to the standard model. We find that in all cases the more advanced logit model structures, which correct for the highly restrictive IID and IIA conditions, provide significantly greater explanatory power than standard logit. Mixed logit and latent class MNL models exhibited the highest overall predictive accuracy on a holdout sample, while the standard logit model performed the worst. Moreover, the analysis of marginal effects of all models indicates that use of advanced models can lead to more insightful and behaviourally meaningful interpretations of the role and influence of explanatory variables and parameter estimates in model estimation. The results of this paper have implications for the use of more optimal logit structures in future research and practice.     

The Taylor Rule and Forecast Intervals for Exchange Rates

In this paper, we examine the Meese–Rogoff puzzle from a different perspective: out‐of‐sample interval forecasting. While most studies in the literature focus on point forecasts, we apply semiparametric interval forecasting to a group of exchange rate models. Forecast intervals for 10 OECD exchange rates are generated and the performance of the empirical exchange rate models are compared with the random walk. Our contribution is twofold. First, we find that in general, exchange rate models generate tighter forecast intervals than the random walk, given that their intervals cover out‐of‐sample exchange rate realizations equally well. Our results suggest a connection between exchange rates and economic fundamentals: economic variables contain information useful in forecasting distributions of exchange rates. We also find that the benchmark Taylor rule model performs better than the monetary, PPP and forward premium models, and its advantages are more pronounced at longer horizons. Second, the bootstrap inference framework proposed in this paper for forecast interval evaluation can be applied in a broader context, such as inflation forecasting.     

汇率弹性化下人民币汇率波动预测分析

本文选取影响人民币汇率波动的有关结构变量,分别通过线性MA模型和基于遗传算法改进的GABP神经网络模型,对人民币汇率波动进行模拟和预测。通过比较发现,汇率缺乏弹性时期,逐月MA模型的历史拟合和样本外预测效果最优;随着汇率改革的不断推进和汇率弹性化的增强,GABP神经网络模型在汇率波动的模拟和预测方面均有最优表现,故汇率波动预测模型应随汇率弹性及其波动特性不同因时制宜。同时结果表明,汇率弹性化能够加深汇率波动及其结构变量间的均衡关系,利率市场化改革应与汇率市场化改革协调推进。     

Reserve overstocking in a highly integrated world. New evidence from Asia and Latin America†

Monthly data are used to investigate reserves management in eight Asian and Latin American countries. Idiosyncratic explanatory variables enter into co-integration relationships based on a stochastic buffer stock model, where a reserve variability measure is obtained via conditional variance approaches. International factors influence the co-integration residuals (representing the excess demands for reserves), which tend to co-move within and across geographical areas. Principal components analysis is then implemented to associate their common drivers with the US fed fund effective interest rate and real-effective exchange rate. This two-step approach sheds light on some controversial aspects of reserves and exchange rate management, such as ‘fear of floating’ and mercantilist behavior. Our results suggest that the size of recent excess reserve holdings is probably overstated.     

Quantile regression estimates and the analysis of structural breaks

The paper considers a test for structural breaks based on quantile regressions instead of OLS estimates. Besides granting robustness, this allows us to verify the impact of a break in more than one point of the conditional distribution. The quantile regression test is then repeatedly implemented as a diagnostic tool to uncover partial or spurious breaks. The test is also implemented to measure the contribution of each explanatory variable to the instability of the regression coefficients, thus finding which one of the different possible sources of breaks linked to the nature of the explanatory variables is the most effective. A real data example of exchange rates shows the presence of a time-driven break, but only at the lower quartile, while the analysis of the explanatory variable excludes its involvement in the break. Since the asymptotic distribution of the OLS test for structural change depends on i.i.d. normal errors and on the exogeneity of the explanatory variables, a Monte Carlo study analyses the behavior of OLS and quantile regression tests for structural changes with lagged endogenous variables, non-normal errors, spurious or partial breaks, and misspecification.     

Addressing unobserved endogeneity bias in accounting studies: control and sensitivity methods by variable type

Together with their associated statistical routines, this paper describes the control and sensitivity methods that can be employed by accounting researchers to address the important issue of unobserved (omitted) variable bias in regression and matching models according to the types of variables employed. As with other social science disciplines, an important and pervasive issue in observational (non-experimental) accounting research is omitted variable bias (endogeneity). Causal inferences for endogenous explanatory variables are biased. This occurs in regression models where an unobserved (confounding) variable is correlated with both the dependent (outcome) variable in a regression model and the causal explanatory (often a selection) variable of interest. The Heckman treatment effect model has been widely employed to control for hidden bias for continuous outcomes and endogenous binary selection variables. However, in accounting studies, limited (categorical) dependent variables are a common feature and endogenous explanatory variables may be other than binary in nature. The purpose of this paper is to provide an overview of contemporary control methods, together with the statistical routines to implement them, which extend the Heckman approach to binary, multinomial, ordinal, count and percentile outcomes and to where endogenous variables take various forms. These contemporary methods aim to improve causal estimates by controlling for hidden bias, though at the price of increased complexity. A simpler approach is to conduct sensitivity analysis. This paper also presents a synopsis of a number of sensitivity techniques and their associated statistical routines which accounting researchers can employ routinely to appraise the vulnerability of causal effects to potential (simulated) unobserved bias when estimated with conventional regression and propensity score matching estimators.     

Demand and supply functions for money in Canada: Some further evidence

The demand and supply functions for money in Canada are estimated in a simultaneous-equation model format in which the supply of money is considered endogenous. The results indicate that the important determinants of the demand for money are income, short-term interest rate and lagged real cash balance variables. Wealth is an unimportant explanatory variable in money demand thus rejecting the hypothesis that an increase in outside wealth increases the demand for money. The monetary base and cash reserve ratio are important explanatory variables in the money supply function and play a direct role in monetary control.     

International order flows: Explaining equity and exchange rate returns

Macroeconomic models of equity and exchange rate returns perform poorly at high frequencies. The proportion of daily returns that these models explain is essentially zero. Instead of relying on macroeconomic determinants, we model equity price and exchange rate behavior based on a concept from microstructure–order flow. The international order flows are derived from belief changes of different investor groups in a two-country setting. We obtain a structural relationship between equity returns, exchange rate returns and their relationship to home and foreign equity market order flow. To test the model we construct daily aggregate order flow data from 800 million equity trades in the U.S. and France from 1999 to 2003. Almost 60% of the daily returns in the S&P100 index are explained jointly by exchange rate returns and aggregate order flows in both markets. As predicted by the model, daily exchange rate returns and order flow into the French market have significant incremental explanatory power for the daily S&P returns. The model implications are also validated for intraday returns.     

Cointegration Analysis of Audit Pricing Model: A Panel Unit Root Test Approach

In this study, we provide evidence on the stationarity of real audit fees and the major explanatory variables frequently used in the audit pricing models from a pooled data set, using panel unit root tests developed by Im et al. (1997). The panel unit root test supports the hypothesis of non‐stationarity of audit fees and their major determinants. We demonstrate that variables in the audit pricing model that were previously found to have impact on audit fees may turn out to be useless when more powerful tests like panel tests are applied to these variables. Our evidence implies that failing to employ appropriate procedure to test cointegration and to specify the appropriate model for audit fees and their determinants would generate results that may have exaggerated the effects of some variables on audit fees.     

投资者关注会影响个体投资者的汇率预期吗?--基于银行代客结售汇数据的实证分析

为考察个体投资者汇率预期的非理性及形成机制,本文基于外汇市场异质主体理论和行为金融学观点,构建了包含投资者关注的个体投资者汇率预期形成模型,并采用远期和即期银行代客结汇售汇比作为个体投资者汇率预期的替代指标,进行了实证分析。结果表明,远期结售比受到前一期汇率变动和境内外利差的影响,而即期结售比受前一期汇率变动影响。进一步引入投资者关注,采用MS-FTP和MS-TVTP模型的实证研究表明,投资者关注对远期和即期结售比的影响是时变的。在人民币贬值阶段,投资者关注对结售比存在显著的负向冲击,会导致更大的结售汇逆差;而在人民币升值阶段,投资者关注对结售比则无显著影响。本文的结论表明,个体投资者汇率预期具有显著的非理性特征,符合动量原则和套息规则,同时还受到投资者关注的影响。央行在人民币处于持续贬值区间时,应通过汇率沟通作用于投资者关注,稳定汇率预期。     

The Association Between Market and Exchange Rate Risks and Accounting Variables: A GARCH Model of the Japanese Banking Institutions

This study has two purposes. First, it estimates the market, interest rate, and exchange rate sensitivities (betas) of the Japanese banking institutions. Second, it investigates the relationship between the market-based measures of risk and accounting-based financial ratios. We extend the literature in three important ways. First, we employ a multi-factor GARCH model to estimate the betas. This framework incorporates non-linearities in the bank stock return modeling and allows for time-varying risk premia. Second, we investigate the determinants of market and exchange rate risk in terms of bank financial ratios. To this end, we regress the beta measures derived from the GARCH model against the corporate decision variables to determine the direction and the magnitude of the impact of the latter on the market and exchange rate risk exposures. Third, by using data on the Japanese banking institutions, we provide a comparison of the bank interest rate and exchange rate sensitivities and the strength of the links between the risk measure and the corporate decision variables between the U.S. and the Japanese banking institutions. This comparison sheds light on the robustness of the results concerning interest rate and exchange rate risk, and their determinants, across the two countries. Several interesting results are obtained. First, empirical results indicate that interest rate is only occasionally significant while market and exchange rate variables are significant for all the banks in the sample. Second, market and exchange rate risk measures do impound information in the financial ratios with the explanatory power of the market beta model being higher than that of the exchange rate beta model. Third, the association of the market-based risk measures and the financial ratios is weaker for the Japanese banks than those found for their U.S. counterparts in the existing literature.JEL Classification: G21, F37     

基于广义线性模型的我国投资型寿险产品退保率实证研究

基于某一款万能险产品38个月的退保率样本数据,结合我国寿险业务的实际情况,利用广义线性模型对影响退保率的主要因素进行了分析。研究结果显示,源于消费者在购买保险的经济环境方面的差异,国外的退保率模型在我国并不适用。从我国实际情况出发,通过选择合适的解释变量,构建了新退保率模型,并分析了各解释变量对退保率的影响。     

VARIABLE TRANSACTIONS COSTS, THE CAPITAL ASSET PRICING MODEL AND THE CORPORATE DIVIDEND DECISION: A COMMENT

This comment addresses some econometric issues raised by Theobald's model of capital market equilibrium in the presence of variable transactions costs. Theil's results on specification error analysis in the absence of relevant explanatory variables are invoked. It is shown that for Theobald's model to be consistent with the empirical findings of Black, Jensen, and Scholes, more stringent requirements must be satisfied.     

Earnings Persistence, Losses and the Estimation of Earnings Response Coefficients

One of the major themes of capital markets accounting research concerns mapping the relation between accounting earnings and security returns. There is still not agreement on the functional form of this relation. The models analysed here are those where: the level of earnings alone, the change in earnings alone, or both, scaled by price, are used as explanatory variables for returns. This article demonstrates that if earnings are either completely permanent or entirely transitory, the earnings response coefficients (ERCs) estimated by levels and changes models should coincide. However, if earnings comprise a mixed process of permanent and transitory components, the ERC estimated by the levels will differ from that estimated by the changes model. Using losses to identify transitory components in earnings, empirical evidence consistent with these predictions is provided.
A combined model using both the level of, and change in, earnings is justified as a weighted average of an earnings and a book value valuation model (e.g., Ohlson, 1989). An alternative rationalization concerns the mitigation of an errors-in-variables problem associated with the estimation of unexpected earnings (Ali and Zarowin, 1992). The results for the combined model are more consistent with the latter. In this context, some previous empirical studies perceive the levels variable as a useful addition to the changes variable when there are transitory components in earnings. However, the evidence reported here suggests that the level of earnings, scaled by price, appears to be the fundamental earnings explanatory variable for returns (Ohlson, 1991, p. 1). The changes variable can, when the errors-in-variables problem is not mitigated by other methods, be a useful addition to the levels variable.     

Changes in the risk structure of stock returns: Consumer Confidence and the dotcom bubble

Changes in the risk structure of stock returns may sometimes be very revealing. We examine economic variables that help explain principal components in UK stock returns, 01/1985 to 12/2001. The loading pattern on explanatory variables for the first component in a ‘bubble’ period is distinctive and consistent with a bubble/crash market. The second component shows a loading pattern on a Consumer Confidence variable in a pre-bubble period only. We observe apparently systematic changes in the structure of risk, and conjecture that Consumer Confidence captures a change in market sentiment that could be a signal for the evolution of stock prices.     

Expected income in the demand for money: An ‘unobservable variable’ analysis

A demand for money function is specified with unobservable expected (or permanent) income as one of the explanatory variables. The model is formulated within the latent or unobservable framework and considers two cases. First, where there is a simple ‘measurement error’ relationship between latent and actual magnitudes, and second where expected income is assumed generated by a more dynamic mechanism. In both cases a consumption function with permanent income as an explanatory variable is incorporated into the model to supply the information necessary to identify the structural parameters. Identification of the models is discussed and a suitable estimation procedure developed and applied to time series data for the United Kingdom.     

Predicting Corporate Financial Distress: A Time-Series CUSUM Methodology

The ability to predict corporate financial distress can be strengthened using models that account for serial correlation in the data, incorporate information from more than one period and include stationary explanatory variables. This paper develops a stationary financial distress model for AMEX and NYSE manufacturing and retailing firms based on the statistical methodology of time-series Cumulative Sums (CUSUM). The model has the ability to distinguish between changes in the financial variables of a firm that are the result of serial correlation and changes that are the result of permanent shifts in the mean structure of the variables due to financial distress. Tests performed show that the model is robust over time and outperforms similar models based on the popular statistical methods of Linear Discriminant Analysis and Logit.     

Exchange rate bubbles: Fundamental value estimation and rational expectations test

We propose a model of periodically collapsing bubbles which extends the Van Norden (1996) model, and nests it, by considering a non-linear specification for the bubble size in the survival regime, and the endogenous determination of the level of the fundamental value of the stochastic process. They allow us to test for rationality in the formation of expectations, and remove the arbitrariness of exogenously setting the level of the fundamental value. This general model is applied to the exchange rate of the Brazilian real to the US dollar from March 1999 to February 2011. The futures market exchange rate is used as a proxy of its expected future value, and three different structural models are considered for the determination of the fundamental value. The first two imply that the exchange rate satisfies either purchasing power parity (PPP), or a modified version of it. The third structural model is a version of the monetary model of exchange rate determination, fitted to the period under consideration. We obtain the maximum likelihood estimate of the parameters of the models, explore the properties of the errors, test its restricted versions, and compare the three specifications for the fundamental. We find that the models we propose fit well the data, and are useful in the heuristic interpretation of the exchange rate movements of the period. Finally we select the structural models that display the best performance, according to several criteria.     

Valuation Model of Defaultable Bond Values in Emerging Markets

This paper develops a model to value defaultable bonds in emerging markets. Default occurs when some signaling process hits a pre-defined default barrier. The signaling variable is considered to be some macro-economic variables such as foreign exchange rates. The dynamics of the default barrier depend on the volatility and the drift of the signaling variable. We derive a closed-form solution of the defaultable bond price from the model as a function of a signaling variable and a short-term interest rate. The numerical results show that the model values generated by using foreign exchange rates as the signaling variables can broadly track the market credit spreads of defaultable bonds in South Korea and Brazil. Given an expected level of the foreign exchange rate, defaultable bond values under a stressed market situation can be obtained.     

Forecasting Corporate Bankruptcy: Optimizing the Performance of the Mixed Logit Model

In recent studies, Jones and Hensher (2004 , 2005) provide an illustration of the usefulness of advanced probability modelling in the prediction of corporate bankruptcies, insolvencies and takeovers. Mixed logit (or random parameter logit) is the most general of these models and appears to have the greatest promise in terms of underlying behavioural realism, desirable econometric properties and overall predictive performance. It suggests a number of empirical considerations relevant to harnessing the maximum potential from this new model (as well as avoiding some of the more obvious pitfalls associated with its use). Using a three-state failure model, the unconditional triangular distribution for random parameters offers the best population-level predictive performance on a hold-out sample. Further, the optimal performance for a mixed logit model arises when a weighted exogenous sample maximum likelihood (WESML) technique is applied in model estimation. Finally, we suggest an approach for testing the stability of mixed logit models by re-estimating a selected model using varying numbers of Halton intelligent draws. Our results have broad application to users seeking to apply more accurate and reliable forecasting methodologies to explain and predict sources of firm financial distress better.