The use of meta-analysis in research on corporate cash holdings speed of adjustment

2019 ◽  
Vol 64 (5) ◽  
pp. 48-73
Author(s):  
Fryderyk Mirota
Keyword(s):  
Selection Bias ◽  
Bayesian Model Averaging ◽  
Meta Analysis ◽  
Estimation Method ◽  
Cash Holdings ◽  
Model Averaging ◽  
Model Specification ◽  
Data Set ◽  
Speed Of Adjustment ◽  
Corporate Cash Holdings

In empirical research significant diversity of corporate cash holdings speed of adjustment (SOA) estimates can be observed. It is possible that some of the results are affected by publication selection bias. Articles whose results are clearly in line with economic theories may be preferred by authors and reviewers and, consequently, conclusions from this area can be published more frequently. The aim of this article is to verify whether there is a publication selection bias with respect to studies related to corporate cash holdings adjustments and to investigate the sources of heterogeneity in cash holdings SOA estimates. The statistical method used in the study was meta-analysis, which allows a combined analysis of the results from independent research. Meta-analysis enables to verify the occurrence of the publication selection bias and to explain the heterogeneity of the results presented in articles. The study was based on data collected asa result of a review of the literature published between 2003 and 2017. On the basis of information on 402 estimates from 58 different studies it has been shown that the publication selection bias does not occur. The Bayesian Model Averaging was used for modelling. It was identified that the characteristics associated with the data set used in the study, model specification and the estimation method significantly affect the hetero-geneity of corporate cash holdings SOA estimates. This diversity is determined, among others, by the choice of estimation method, length of the period covered by the analysis and characteristics of the market environment of the concerned entities.

How to improve allometric equations to estimate forest biomass stocks? Some hints from a central African forest

2014 ◽  
Vol 44 (7) ◽  
pp. 685-691 ◽  
Author(s):  
Quentin Moundounga Mavouroulou ◽  
Alfred Ngomanda ◽  
Nestor Laurier Engone Obiang ◽  
Judicaël Lebamba ◽  
Hugues Gomat ◽  
...  
Keyword(s):  
Measurement Error ◽  
Bayesian Model Averaging ◽  
Sampling Error ◽  
Forest Biomass ◽  
Model Averaging ◽  
Allometric Equation ◽  
Allometric Equations ◽  
Data Set ◽  
Equation Error ◽  
Source Of Error

Predicting the biomass of a forest stand using forest inventory data and allometric equations involves a chain of propagation of errors going from the sampling error to the tree measurement error. Using a biomass data set of 101 trees in a tropical rain forest in Gabon, we compared two sources of error: the error due to the choice of allometric equation, assessed using Bayesian model averaging, and the biomass measurement error when tree biomass is calculated from tree volume rather than directly weighed. Differences between allometric equations resulted in a between-equation error of about 0.245 for log-transformed biomass compared with a residual within-equation error of 0.297. Because the residual error is leveled off when randomly accumulating trees whereas the between-equation error is incompressible, the latter turned out to be a major source of error at the scale of a 1 ha plot. Measuring volumes rather than masses resulted in an error of 0.241 for log-transformed biomass and an average overestimation of the biomass by 19%. These results confirmed the choice of the allometric equation as a major source of error but unexpectedly showed that measuring volumes could seriously bias biomass estimates.

scholarly journals Comparative Risk Assessment for Fossil Energy Chains Using Bayesian Model Averaging

Energies ◽  
2020 ◽  
Vol 13 (2) ◽  
pp. 295
Author(s):  
Matteo Spada ◽  
Peter Burgherr
Keyword(s):  
Bayesian Model ◽  
Bayesian Model Averaging ◽  
Model Averaging ◽  
Risk Indicators ◽  
Severe Accident ◽  
Accident Risk ◽  
The European Union ◽  
Data Set ◽  
Fossil Energy ◽  
Accident Data

The accident risk of severe (≥5 fatalities) accidents in fossil energy chains (Coal, Oil and Natural Gas) is analyzed. The full chain risk is assessed for Organization for Economic Co-operation and Development (OECD), 28 Member States of the European Union (EU28) and non-OECD countries. Furthermore, for Coal, Chinese data are analysed separately for three different periods, i.e., 1994–1999, 2000–2008 and 2009–2016, due to different data sources, and highly incomplete data prior to 1994. A Bayesian Model Averaging (BMA) is applied to investigate the risk and associated uncertainties of a comprehensive accident data set from the Paul Scherrer Institute’s ENergy-related Severe Accident Database (ENSAD). By means of BMA, frequency and severity distributions were established, and a final posterior distribution including model uncertainty is constructed by a weighted combination of the different models. The proposed approach, by dealing with lack of data and lack of knowledge, allows for a general reduction of the uncertainty in the calculated risk indicators, which is beneficial for informed decision-making strategies under uncertainty.

The Speed of Adjustment of Corporate Cash Holdings

2018 ◽  
Vol 52 (3) ◽  
pp. 139-151 ◽  
Author(s):  
K T Vigneswara Rao ◽  
Keyur Thaker*
Keyword(s):  
Cash Holdings ◽  
Speed Of Adjustment ◽  
Corporate Cash Holdings

scholarly journals A Bayesian Model-Averaged Meta-Analysis of the Power Pose Effect with Informed and Default Priors: The Case of Felt Power

2017 ◽  
Author(s):  
Quentin Frederik Gronau ◽  
Sara van Erp ◽  
Daniel W. Heck ◽  
Joseph Cesario ◽  
Kai Jonas ◽  
...  
Keyword(s):  
Bayesian Model ◽  
Bayesian Model Averaging ◽  
Meta Analysis ◽  
Random Effect ◽  
Random Effect Model ◽  
Model Averaging ◽  
Fixed Effect Model ◽  
Body Postures ◽  
Effect Model ◽  
Meta Analyses

Carney, Cuddy, and Yap (2010) found that --compared to participants who adopted constrictive body postures-- participants who adopted expansive body postures reported feeling more powerful, showed an increase in testosterone and a decrease in cortisol, and displayed anincreased tolerance for risk. However, these power pose effects have recently come under considerable scrutiny. Here we present a Bayesian meta-analysis of six preregistered studies from this special issue, focusing on the effect of power posing on felt power. Our analysisimproves on standard classical meta-analyses in several ways. First and foremost, we considered only preregistered studies, eliminating concerns about publication bias. Second, the Bayesian approach enables us to quantify evidence for both the alternative and the null hypothesis. Third, we use Bayesian model-averaging to account for the uncertainty with respect to the choice for a fixed-effect model or a random-effect model. Fourth, based on a literature review we obtained an empirically informed prior distribution for the between-studyheterogeneity of effect sizes. This empirically informed prior can serve as a default choice not only for the investigation of the power pose effect, but for effects in the field of psychology more generally. For effect size, we considered a default and an informed prior. Our meta-analysis yields very strong evidence for an effect of power posing on felt power. However, when the analysis is restricted to participants unfamiliar with the effect, the meta-analysis yields evidence that is only moderate.

scholarly journals Meta-Analysis with Few Studies and Binary Data: A Bayesian Model Averaging Approach

Mathematics ◽  
2020 ◽  
Vol 8 (12) ◽  
pp. 2159
Author(s):  
Francisco-José Vázquez-Polo ◽  
Miguel-Ángel Negrín-Hernández ◽  
María Martel-Escobar
Keyword(s):  
Model Uncertainty ◽  
Binary Data ◽  
Posterior Probability ◽  
Bayesian Model Averaging ◽  
Meta Analysis ◽  
Cluster Structure ◽  
Model Averaging ◽  
Cluster Models ◽  
Mixing Distribution ◽  
Sample Heterogeneity

In meta-analysis, the existence of between-sample heterogeneity introduces model uncertainty, which must be incorporated into the inference. We argue that an alternative way to measure this heterogeneity is by clustering the samples and then determining the posterior probability of the cluster models. The meta-inference is obtained as a mixture of all the meta-inferences for the cluster models, where the mixing distribution is the posterior model probabilities. When there are few studies, the number of cluster configurations is manageable, and the meta-inferences can be drawn with BMA techniques. Although this topic has been relatively neglected in the meta-analysis literature, the inference thus obtained accurately reflects the cluster structure of the samples used. In this paper, illustrative examples are given and analysed, using real binary data.

Determinants of corruption in developing countries: case of Tunisia

2021 ◽  
Vol ahead-of-print (ahead-of-print) ◽  
Author(s):  
Azzouz Zouaoui ◽  
Mounira Ben Arab ◽  
Ahmad Mohammed Alamri
Keyword(s):  
Developing Countries ◽  
Unemployment Rate ◽  
Bayesian Model Averaging ◽  
Estimation Method ◽  
Model Averaging ◽  
Public Expenditures ◽  
Annual Data ◽  
Content Type ◽  
Determinants Of Corruption ◽  
The Government

Purpose This paper aims to investigate the economic, political or sociocultural determinants of corruption in Tunisia. Design/methodology/approach To better understand the main determinants of corruption in Tunisia. This study uses The Bayesian Model Averaging (BMA) model, which allows us to include a large number of explanatory variables and for a shorter period. Findings The results show that economic freedom is the most important variable of corruption in Tunisia. In second place comes the subsidies granted by the government, which is one of the best shelters of corruption in Tunisia through their use for purposes different from those already allocated to them. Third, this paper finds the high unemployment rate, which, in turn, is getting worse even nowadays. The other three factors considered as causal but of lesser importance are public expenditures, the human development index (HDI) and education. Education, the HDI and the unemployment rate are all socio-economic factors that promote corruption. Originality/value The realization of this study will lead to triple net contributions. The first is to introduce explicitly and simultaneously political, social and economic determinants of corruption in developing countries. Second, unlike previous studies based on the simple and generalized regression model, the present research uses another novel and highly developed estimation method. More precisely, this study uses the BMA model, on the set of annual data for a period of 1998–2018. The third contribution of this research resides in the choice of the sample.

Distinguishing Normative Processes From Noise

2016 ◽  
Vol 79 (4) ◽  
pp. 311-332 ◽  
Author(s):  
Jonathan H. Morgan ◽  
Kimberly B. Rogers ◽  
Mao Hu
Keyword(s):  
Bayesian Methods ◽  
Bayesian Model ◽  
Stepwise Regression ◽  
Bayesian Model Averaging ◽  
Model Averaging ◽  
Ground Truth ◽  
Model Specification ◽  
Affect Control Theory ◽  
False Discovery ◽  
Social Events

This research evaluates the relative merits of two established and two newly proposed methods for modeling impressions of social events: stepwise regression, ANOVA, Bayesian model averaging, and Bayesian model sampling. Models generated with each method are compared against a ground truth model to assess performance at variable selection and coefficient estimation. We also assess the theoretical impacts of different modeling choices. Results show that the ANOVA procedure has a significantly lower false discovery rate than stepwise regression, whereas Bayesian methods exhibit higher true positive rates and comparable false discovery rates to ANOVA. Bayesian methods also generate coefficient estimates with less bias and variance than either stepwise regression or ANOVA. We recommend the use of Bayesian methods for model specification in affect control theory.

Sign in / Sign up

Export Citation Format

Share Document