Merger and acquisitions in South African banking: A network DEA model

Banking in South Africa is known for its small number of companies that operate as an oligopoly. This paper presents a strategic fit assessment of mergers and acquisitions (M&A) in South African banks. A network DEA (Data Envelopment Analysis) approach is adopted to compute the impact of contextual variables on several types of efficiency scores of the resulting virtual merged banks: global (merger), technical (learning), harmony (scope), and scale (size) efficiencies. The impact of contextual variables related to the origin of the bank and its type is tested by means of a set of several robust regressions to handle dependent variables bounded in 0 and 1: Tobit, Simplex, and Beta. The results reveal that bank type and origin impact virtual efficiency levels. However, the findings also show that harmony and scale effects are negligible due to the oligopolistic structure of banking in South Africa.     

Assessing impact of regulatory risk on shareholders’ wealth

Research indicates that regulatory risk increases required return on investment by investors and causes underinvestment in industries with high sunk costs. The effects of regulatory changes may be measured by estimating the abnormal returns associated with the event. The results may suggest to regulators what should be encouraged or avoided. This article utilizes a fixed effects regression to examine abnormal returns from changes in Philippine nationalization regulations. The results are consistent with extant literature. Supreme Court decisions, which increased uncertainty and regulatory risk, produced negative abnormal returns. The initial release of draft implementing rules did not produce statistically significant effects, but a succeeding draft favouring liberalization, produced positive abnormal returns.     

新型农村社会养老保险对贫困群体生活的影响——基于 CHARLS2011～2015面板数据的分析

本文探讨了新型农村社会养老保险是否对低收入的贫困群体的生活产生更显著的效用,从而论证该制度是否助力扶贫工作。因为养老金是收入的组成部分,而收入有边际效用递减的特点,因此养老金对不同收入群体的效用可能存在差别。本文按家庭人均收入水平区分贫困与非贫困群体,基于CHARLS 2011 -2015年数据,采用多维贫困指标,使用断点回归发现新农保养老金对贫困参保群体在食品消费方面产生更显著的效用,有助于扶贫工作。并使用含复杂时变效应的二元变量模型,进一步发现该效用在养老金开始发放时显现。本文结论有利于认识新农保政策对扶贫的作用,并将扶贫工作和已有的社会保险政策结合,以提高扶贫工作效率。     

提升算法对传统车险索赔频率建模模型的改进——基于我国五省交强险保单数据

广义线性模型作为非寿险定价的经典模型,在非寿险定价中得到了广泛的应用。近年来,以提升算法为代表的机器学习算法在保险领域取得了很好的效果,为保险产品定价提供了一种新的选择。本文将提升算法思想分别融入到回归树模型和广义线性模型(GLM)中去,用得到的新模型对我国车险索赔频率进行预测建模分析,并与传统的回归树模型和GLM进行比较。结果表明,加入提升算法后传统车险索赔频率建模模型的效果得到了很大的改善,并且在不存在过拟合的前提下,随着模型深度和迭代次数的增加,模型的效果也在不断优化。     

On Minimally-supported <Emphasis Type="Italic">D</Emphasis>-optimal Designs for Polynomial Regression with Log-concave Weight Function

This paper studies minimally-supported D-optimal designs for polynomial regression model with logarithmically concave (log-concave) weight functions. Many commonly used weight functions in the design literature are log-concave. For example, and exp(−x ²) in Theorem 2.3.2 of Fedorov (Theory of optimal experiments, 1972) are all log-concave. We show that the determinant of information matrix of minimally-supported design is a log-concave function of ordered support points and the D-optimal design is unique. Therefore, the numerically D-optimal designs can be constructed efficiently by cyclic exchange algorithm.     

Which Elasticity? Estimating the Responsiveness of Taxpayer Reporting Decisions

In this paper we present estimates of the responses of individuals to marginal tax rates in their reporting of income, using data from individual tax returns for the year 1995. One estimation method is ordinary least squares regression. A second method uses quantile regression, which provides evidence on behavioral responses at different points (or quantiles) in the distribution of income and so is relevant to the question of whether the responses of, say, the rich differ from those at other points in the income distribution. Our results clearly indicate that marginal tax rates affect the reporting decisions of individuals. However, there are significant differences in the marginal tax rate reporting responses for the various types of reported income, there are major differences across income classes, and there are notable differences in the estimated responses across estimation methods.

投资者关注对金银期货市场收益影响的时频分位研究

根据Google投资者关注度指数和金银期货市场交易数据,构建基于小波分解序列的时频门限自回归分布滞后模型,通过分位数模型参数估计,基于时域与频域联合分析视角,考量投资者关注度对金银期货市场收益的影响。结果表明：投资者关注度对金银期货市场的影响具有异质性;在低频域内,投资者关注度对金银期货市场影响相对较小;极端分位数水平下,投资者关注度对金银期货市场收益影响的时效性较短,投资者关注度对白银期货市场收益的影响较弱。     

我国农用化肥施用的影响因素——基于省际面板数据的实证分析

我国农用化肥施用量不断提高带来的环境污染问题日益严重。研究化肥施用的影响因素及其影响程度,为防止化肥滥施、保护生态环境提供依据意义重大。文章采用1999～2008年我国31个省(市、区)的面板数据,通过面板单位根检验、协整关系检验和多元回归方法分别对国内东、中、西三大区域化肥施用密度与其影响因素之间的关系进行了实证分析。研究结果表明,三大区域农民人均纯收入、人口规模对化肥施用密度具有显著的正向影响,三大区域化肥价格增长率、粮食种植面积比重、化肥产出效率对化肥施用密度具有显著的负向影响,三大区域种植产品生产价格、种植业劳动力数量、农产品国际贸易对化肥施用密度具有不同的影响效果或显著性。     

河北省农村基础设施建设的收入效应研究

运用1995～2008年的数据,采用ADF检验、协整检验等对河北省农民人均纯收入和农村用电量、农业基本建设投资额及农民家庭平均每户的交通运输、仓储和邮政业原值进行了实证分析。结果表明三者对农村经济增长具有不同程度的推动作用。政府应该通过明确农村基础设施供给中各级政府的职责,加大农村基础设施尤其是生产性基础设施的投资力度等措施来促进农村经济的发展。     

云南省城镇居民消费与收入关系的非参数回归分析

近二十多年来,非参数统计的方法已广泛应用于计量经济学的研究。由于不需要预先确定变量间的函数关系,对计量经济模型的估计和预测提供了多方面和灵活机动的表现形式。本文尝试将非参数估计理论引入到回归模型中,通过建立非参数回归模型及普通的线性回归模型对云南省城镇居民可支配收入及消费之间的关系进行比较研究。