2005). Meta-regressionanalysis (MRA) is a type of meta-analysis that objectively explains why and quantifies how estimates from a range of empirical studies differ (Roberts 2005). MRA provides a framework for replicating results from different studies and offers a sensitivity analysis for model specification (Stanley 2005). Its intent is to summarize the results of many individual studies, where key estimates differ in significance, magnitude and even sign. MRA provides a more general description of the relationship between the variables, and can identify a significant trend from a large number of studies, even where individual studies might fail to find such evidence (Mann 1990, 1994).
Based on the findings of meta-regressionanalysis, this paper does a empirical revaluation using all the A shares from 2008 to 2014 aimed at exploring the impact of Chinese corporate size on corporate tax avoidance. The results of Hausman tests indicate that fixed-effect models are superior to random-effect models. So this paper sets up four fixed-effect models. The four explained variables are CETR, CTS, BTD and DDBTD.
Abstract: The main focus of this paper is to survey the literature that investigates the effects of exchange rate uncertainty on international trade. Specifically, we carry out meta-regressionanalysis to 42 studies with 810 estimates. We show that the empirical studies on the focal link exhibit a substantial publication selection and a significant genuine exchange rate volatility effect on trade flows after correction of publication bias. Moreover, we find that most of the variables that may help explain the heterogeneity of results (such as the country sample, the choice of modeling strategies and potential political measures, etc) are significant. These results appear robust among the different methods used and the dummies for the type of research outlet and the publication year included in our estimates.
coefficient suggested that real estate assets had a greater effect on large capitalisation stocks than small capitalisation stocks. Importantly for investors who may be trying to diversify assets, the findings indicate that the effect of the stock market on real estate has systematically reduced over time. The number of observation is found to reduce the observed effect, perhaps flagging that the more reliable results find a smaller relationship. Where the study uses cross section or panel data, the reported effect of the stock market on real estate is greater. This suggests that on an aggregated level the effect of the stock market on real estate is greater than when individual markets are considered in isolation. Other methodologies used by studies did not appear to significantly affect the results. Unlike the previous analysis, the second meta-regressionanalysis did not find the number of authors to have a significant affect on the results.
The data used for this study are obtained from the Environmental Valuation Information System (EVIS) provided by Korea Environment Institute (KEI). The EVIS establishes the database associated with environmental valuation studies conducted in South Korea and to form the relevant studies we focused on the previous studies which measured the WTPs with respect to water quality improvement in South Korea. As mentioned earlier, since the four major rivers in Korea have provided stakeholders with various benefits such as a primary source of drinking water, recreational activities, and aesthetic amenities, this study aims to estimate total use value which consists of direct and indirect use value. Included are, therefore, all eligible studies in the meta-regressionanalysis which estimated economic values for consumptive use and/or non-consumptive use of water resources in South Korea. Table 1. Definition and relationship between water quality grade and water pollution index.
Meta-regressionanalysis will be used to systematically explore the reasons for different effect sizes and out- comes across studies. Meta-regressions are similar to standard regressions, in which an outcome variable is predicted according to the values of one or more ex- planatory variables. In meta-regression, the dependent variable is the effect estimate (in this case, the SMD for each study). The independent or explanatory variables are characteristics of studies that might influence the size of the intervention effect. In meta-regression, larger studies have more influence on the relationship than smaller studies, since studies are weighted by the preci- sion of their respective effect estimate. We will use a random-effects meta-regression in which the residual heterogeneity among intervention effects not captured by the independent explanatory variables is incorporated in the same way as in a random effects meta-analysis.
Economic theory and empirical evidence indicate that technological innova- tion is an important determinant of long-term economic development. Vari- ous country policies have been launched in favour of private research and development (R&D) with economic development as the main objective. As often in economics, public intervention is grounded on the presumed exis- tence of market failures. The purpose of this paper is two-fold. First, it pro- vides an overview of the history of R&D-related tax policies in more than ten industrial countries. Second, after reviewing the existent empirical evidence on the effectiveness of R&D tax credits policies, it presents a meta-regressionanalysis based on an econometric model. Our results show that an R&D tax credit is strongly significant in the studies taken cumulatively.
Metformin is a common anti-diabetic drug with both systemic and cardioprotective benefits in addition to its hypoglycaemic effect [9, 10]. At the cellular level met- formin activates adenosine monophosphate-activated protein kinase (AMPK) an important regulator of several metabolic pathways resulting in enhanced glucose utili- sation, reduction of protein synthesis and improvement of mitochondrial function [11–13]. Furthermore, met- formin has been shown to reduce collagen accumulation and potentially reduce LV hypertrophy and improve dias- tolic function in the diabetic myocardium . Several observational series have shown a reduction in mortal- ity in the HF population [15, 16]. Its mortality benefit in the HFpEF population however has yet to be explored. We performed a systematic review and meta-regressionanalysis to identify whether variations in ejection fraction (EF) impact mortality outcomes in HF patients treated with metformin.
Finding definitive results in meta-regressionanalysis requires that all the factors characterizing individual studies be included in the model. Furthermore, the linearity of the model (assumed in most analyses of multiple regression-based econometric studies) ignores possible non-linearity in the relationships among the explanatory variables. Nevertheless, periodic examination of empirical studies provides a systematic check on results to date. As new databases become available, and additional variables are included in cost and production function models, our understanding of causal links between inputs and outputs improves. Here, the focus has been on economies of scale and scope, topics that are central to decisions to consolidate current operators or to decentralize operations. However, the decision-relevance of the scholarly literature is still an open question. Is it obvious that a utility operating in a region of diseconomies of scale should be split up, as in the case of Manila? Were the resulting performance gains in Manila due to the availability of yardstick comparisons, privatization, or to achieving an “optimal” size? Should two separate organizations be maintained for water and wastewater when a coefficient of a study suggests that economies of scope are being missed? Although few would argue against “evidence-based” decision-making, the weight given to facts is unlikely to be the deciding factor for those responsible for water sector policies.
public health priority aiming to reduce the burden of obesity and co-morbidities. Therefore this review aims to assess the effectiveness of non-surgical weight management interventions for ob- esity in the UK. Method: Thirty one databases were searched that identified 20 articles for inclu- sion. Articles were screened and quality scored using the “Effective Public Health Practice Project Quality Assessment Tool”. Meta-regressionanalysis (MRA) was undertaken on seven studies that allowed for effect size calculations. Results: In adult populations, lifestyle interventions that tar- geted both diet and physical activity, delivered in the private sector were most effective in reduc- ing weight and/or BMI, and were more cost-effective. In children the most successful interventions mirrored adult interventions, but were family-orientated. MRA supported these findings. Most frequent intervention duration was 12 weeks. Discussion: The results provide evidence to support policy makers for the effective delivery of weight management interventions. Findings suggest that weight management interventions in the UK are effective in reducing weight and/or BMI for both children and adults. Interventions delivered in the private sector (e.g. Weight Watchers), targeting diet and physical activity levels, demonstrated the highest levels of effectiveness. How- ever, compared to these models, NHS programmes are less well defined in the research literature
We conduct a hierarchical meta-regressionanalysis to review 87 empirical studies that report 769 estimates for the effects of government size on economic growth. We follow best- practice recommendations for meta-analysis of economics research, and address issues of publication selection bias and heterogeneity. When size is measured as the ratio of total government expenditures to GDP, the partial correlation between government size and per- capita GDP growth is negative in developed countries, but insignificant in developing countries. When size is measured as the ratio of consumption expenditures to GDP, the partial correlation is negative in both developed and developing countries, but the effect in developing countries is less adverse. We also report that government size is associated with less adverse effects when primary studies control for endogeneity and are published in journals and more recently, but it is associated with more adverse effects when primary studies use cross-section data. Our findings indicate that the relationship between government size and per-capita GDP growth is context-specific and likely to be biased due to endogeneity between the level of per-capita income and government expenditures.
We aim to address these issues through meta-regressionanalysis, a quantitative method of literature review that has been used extensively in medical research and has gathered momentum in economics research (Stanley and Doucouliagos, 2012; Stanely et al., 2013). Focusing on primary studies informed by a derived labour demand model (DLDM), we report the following findings: (i) the extent of between-study heterogeneity that cannot be explained by sampling variations is high (over 75%) in the full sample and in some of the sub-samples that reflect specific combinations of innovation and skill types; (ii) the effect-size is positive but small in the full sample and in subsamples that capture different combinations of innovation and skill types; (iii) the effect on the demand for unskilled labour is smaller than skilled or mix- skills labour demand, but there is no evidence of negative effect on unskilled labour demand; (iv) there is evidence of moderate positive publication selection bias in the overall evidence base, but the bias is large and reflects selection in favour of theoretical predictions in the case of process and product innovation subsamples; (v) the evidence based on firm/industry data from six OECD countries reveals a U-shaped relationship between the ‘effect-size’ estimates and labour/product market regulation; and (vii) although the effect is larger in primary studies published after 2000, it is relatively smaller when the primary studies use panel data and instrumental variable estimation methods, draw on data related to high-innovation-intensity firms/industries, and they measure innovation with intellectual property assets.
using 12 mg or less of haloperidol with those that used a higher dose. In the trials in which the mean haloperi- dol dose was <12mg/day, the random effects standardised weighted mean difference was − 0.09 ( − 0.07 to − 0.26), whereas in those with a mean haloperidol dose of > 12 mg/day it was − 0.28 ( − 0.13 to − 0.44) (fig 1). There was no difference in dropout rates between atypical antipsychotics and haloperidol in the trials that used <12 mg/day haloperidol (pooled risk difference was − 0.1% ( − 4.6% to 4.4%) with the random effects model, but the pooled drop out in trials using > 12 mg/day haloperidol was − 8.3% ( − 1.3% to 15.2%) (fig 2). In other words, the advantages of atypi- cal antipsychotics in terms of efficacy and dropout rates are not seen if haloperidol is used at doses of 12mg/day or less. These results from the meta- regressionanalysis were unaffected by the removal of trials including treatment resistant patients, those taking sertindole, or long term trials (data not shown). We examined whether the lower incidence of extrapyramidal side effects with atypical antipsychotics was dose related. The trials used a range of measures to describe side effects, making meta-analysis and meta-
designed, adequately powered, head-to-head clinical tri- als. As the results of placebo-controlled trials are often suf- ficient to acquire the regulatory approval of new drugs, pharmaceutical companies may not be motivated to sup- port trials that compare new drugs with existing active treatments. Lack of evidence from direct comparison between active interventions makes it difficult for clini- cians to choose the most effective treatment for patients . Because of the lack of direct evidence, indirect com- parisons have been recommended . Adjusted indirect comparison is a way to compare two compounds through their relative effect vs. a common comparator (placebo in our study). The indirect approach to meta-analysis requires certain conditions to yield optimal results. Differ- ences in study designs, inclusion/exclusion criteria, patients characteristics at baseline as well as difference in drug dosage  and publication bias are limitations that may lead to unbalanced conclusions  and merit dis- cussion.
Finally, there are many considerations to take into ac- count when ensuring our review has sufficient power. From our initial pilot search (Additional file 2), we un- covered a vast amount of literature examining celebrity impacts on different health topics such as smoking, body image, suicide, and cancer screening. In order to effectively analyze such a heterogeneous pool of data and provide meaningful conclusions, we must carefully categorize all included studies by themes and outcomes before performing any statistical tests. This has been addressed by conducting data extraction in three dis- tinct stages, with the first two stages analyzing the study designs and outcome measures in order to group them before proceeding with the final phase of data ex- traction. While organizing the studies in this way is crucial to achieving a meaningful analysis, it may de- crease the number of studies available for statistical testing with respect to each outcome. However, by encompassing a wide range of quantitative and qualita- tive evidence in our study, we aim to produce meaning- ful findings to enhance our understanding of the present celebrity-health phenomenon and guide policy- makers in facilitating significant and positive public health initiatives.
Several important conclusions can be drawn from this analysis. As already acknowledged in the literature, the ETI is not a structural parameter and this study shows that policy conclusions can be misleading. Reported estimates need to be interpreted within the context they are estimated in and researchers and policy makers need to be careful about what type and size of elasticity should be used for policy analysis (e.g. when calibrating an optimal tax model). Moreover, the analysis shows that results are sensitive. The choice of income control and difference length always affects estimation results. It is remarkable how many details about the (constructed) income measures are missing in primary studies (e.g. what is included in the tax base). Following Kopczuk (2015) „sensitivity of reported estimates is due to model mis-specification, lack of credible variation and poor understanding of the data“. I agree and argue that future research should pay more attention to income dynamics and the tax reform used for identification. For instance, questions like who is affected and to what extent income grows irrespective of the reform should be analyzed more carefully. Based on this, one may find a strategy that produce robust results. I argue that not only more descriptive evidence might be helpful, but more transparency with respect to the underlying tax simulation model is essential. Moreover, instead of proving a single estimate, a range of estimates is more meaningful in order to shed light into the heterogeneity of behavioral responses across the income distribution and different socioeconomic groups.
GlaxoSmithKline Inc. funded this research and was involved in all stages of study conduct, including analysis of the data. GlaxoSmithKline Biologicals SA took in charge all costs associated with the development and publication of this manuscript. We would like to acknowledge Rohita Sharma and Paul Grootendorst with their help with the protocol development for this study. The authors also thank Heather Santiago (publications manager, GSK) and Grégory Leroux (publications manager, Business & Decision Life Sciences on behalf of GSK Vaccines) for editorial assistance and manuscript coordination and Jemila Hamid for her assistance in producing the figures. ACT acknowledges support from a Canadian Institutes of Health Research/Drug Safety and Effectiveness Network New Investigator Award in Knowledge Synthesis.
In this systematic analysis, 87 RCTs were included to evaluate the reliability of different SOFA derivatives to predict treatment effects on mortality. Based on study level data aggregated in this systematic review, Delta fixed-day SOFA appears to be most responsively and consistently associated with mortality. Fixed-day SOFA was the most frequently reported outcome measure among the reviewed RCTs but was not found to be asso- ciated with mortality. Maximum SOFA showed excellent responsiveness and consistency, but was used in too few trials for sufficient statistical power. We recommend that researchers planning to use SOFA as a trial end- point should use Delta SOFA in preference to Fixed-day SOFA, choose an appropriate timeframe, describe how discharged and deceased patients are scored and evalu- ate the within-trial association between the SOFA end- point and mortality.
of the dependent variable, (ii) the type of the fiscal rule, (iii) the administrative level the primary study’s data is referring to, (iv) a general set of control variables usually employed in studies within this strand of literature, (v) indicators testing for a potential publication bias, and (vi) characteristics of the econometric specification. Additionally, we pay particular attention to strategies used to deal with the issue of endogeneity and subsume respective strategies in the group (vii) identification strategies. Within these groups, usually several individual variables account for different aspects of the respective topic. In order to test for the robustness of our results, we extend the set of groups and include the country as well as time coverage of primary studies. A detailed description of all meta-analytical variables is provided in Table A.2 in the appendix.
Regression analyses revealed several predictors with a statistically discernable relationship with costs. In each country, higher service volume was strongly associated with lower average costs, with output elasticity ranging from 1.27 to 2.52 across the six countries, with a mean value of 1.63, and the largest 20% of sites in each country had a cost per dose that was on average 61% lower than that observed for the smallest 20% of sites. For most coun- tries there were many small sites with high average costs, and these small sites exhibited substantial variation in unit costs. The reduction in unit costs associated with higher service volume was only minimal for sites at the upper end of the service volume distribution, and these large sites exhibited only minor variation in unit costs. While this suggests that greater reductions in average costs might be achieved through efforts to improve efficiency in small sites, it is not clear that these sites should be priori- tized, as large total cost reductions might be possible with only small reductions in unit costs at large sites.