A comparison of model choice strategies for logistic regression

Purpose:The purpose of this study is to develop and compare model choice strategies in context of logistic regression.Model choice means the choice of the covariates to be included in the model.Design/methodology/approach:The study is based on Monte Carlo simulations.The methods are compared in terms of three measures of accuracy:specificity and two kinds of sensitivity.A loss function combining sensitivity and specificity is introduced and used for a final comparison.Findings:The choice of method depends on how much the users emphasize sensitivity against specificity.It also depends on the sample size.For a typical logistic regression setting with a moderate sample size and a small to moderate effect size,either BIC,BICc or Lasso seems to be optimal.Research limitations:Numerical simulations cannot cover the whole range of data-generating processes occurring with real-world data.Thus,more simulations are needed.Practical implications:Researchers can refer to these results if they believe that their data-generating process is somewhat similar to some of the scenarios presented in this paper.Alternatively,they could run their own simulations and calculate the loss function.Originality/value:This is a systematic comparison of model choice algorithms and heuristics in context of logistic regression.The distinction between two types of sensitivity and a comparison based on a loss function are methodological novelties.

Users’Participation in Community Forest Management:A Comparative Study

Local community participation in forest management is pivotal since they are familiar with the forest environment.In the successful management of community forestry(CF),both males and females along with the representation of poor and disadvantaged groups are of vital importance.This research compares the users’perception in community forest management(CFM)activities,and socio-economic variables influencing participation in studied community forestry user groups(CFUGs).Primary data were collected through reconnaissance surveys,interviewing key informants,focus group discussions,and household surveys.Secondary data were collected from the division forest office,CFUGs’operational plan(OP)and Constitution,internet,and authenticated websites.The chi-square(χ^(2))test was applied to test separately association variables like gender,caste,age class,education level,and wealth ranking with participation.Using ordered logit regression,the variables affecting participation in OP and constitution-making,Silvicultural activities,Forest products collection,and CF fund mobilization were quantified.Gender and Education were found to be the most promising factor influencing participation in Jagriti CFUG and Jhankrikhola CFUG respectively.In general,higher caste,older age,and rich people dominate the major decision-making activities.However,lower caste and poor people have been involved comparatively more in Forest product collection.

A socioeconomic approach to the profile of microcredit holders from the Hispanic minority in the USA

The subject of this study is the microcredit market in the USA,more specifically in Florida.The justification for choosing this specific state is the massive presence of the Hispanic population.This will facilitate a generalization of the obtained results to the microcredit market in Latin American countries.Thus,the objective of this study is to analyze the profile of microcredit holders and their companies from socioeconomic and financial points of view.As our data also consider the degree of repayment of the microloans included in the sample,the clients’profile is related to the punctuality or default of their corresponding loan repayments using the methodology of multi-nomial logit regression.The variables used in this study refer to personal information concerning borrowers(gender,age,education level,and marital status),the economic situation of their respective companies(closeness to the lender,number of workers,and revenues),and the characteristics of granted loans(principal,term,and purpose).However,the results of the regression show that only two variables are significant at the 5%significance level:the borrower’s age,which has a positive effect on repay-ment punctuality,and the loan term,which exhibits a negative effect.The findings of this study have clear implications,as they can help lenders design suitable microloans adjusted to customer profiles.Finally,future research should include other demograph-ics and characteristics of affected companies.

Multivariate Analyses for Finding Significant Track Irregularities to Generate an Optimal Track Maintenance Schedule

We first discuss the relationship between the optimal track maintenance scheduling model and an efficient detection method for abnormal track irregularities given by the longitudinal level irregularity displacement (LLID). The results of applying the cluster analysis technique to the sampling data showed that maintenance operation is required for approximately 10% of the total lots, and these lots were further classified into three groups according to the degree of maintenance need. To analyze the background factors for detecting abnormal LLID lots, a principal component analysis was performed;the results showed that the first principal component represents LLIDs from the viewpoints of the rail structure, equipment, and operating conditions. Binomial and ordinal logit regression models (LRMs) were used to quantitatively investigate the determinants of abnormal LLIDs. Binomial LRM was used to characterize the abnormal LLIDs, whereas ordinal LRM was used to distinguish the degree of influence of factors that are considered to have a significant impact on LLIDs.

Valuation of the External Cost Caused by the Environmental Pollution of Three Lakes in Northern Greece

The preservation/restoration of natural environment is usually entailing high cost mostly paid by citizens through taxes.The effect of these taxes is double. The direct effect is the obvious additional income for the State, and the indirect effect is anadditional income for the citizen, due to increasing tourism. Since the evaluation of this good cannot be in market terms, authorsapply a modified CVM （Contingent Valuation Method）, which is part of Experimental Economics, in order to find out the order ofconcern that people have about natural environment. Authors also, try to investigate their WTP （Willingness To Pay） for supportingactivities for preservation/restoration of three lakes in Northern Greece, in particular, lake of Ioannina, lake of Florina and lake ofKastoria. For the purpose of this research, authors use parametric and non-parametric approaches, as well as Linear Regression andLogic Models.

Local people’s knowledge on the adverse impacts and their attitudes towards growing Eucalyptus woodlot in Gudo Beret Kebele, Basona Worena district, Ethiopia

Introduction:Growing Eucalyptus at a farm level in the form of woodlot has become popular among rural households in Ethiopia.For example,rural households mainly establish Eucalyptus woodlot as a component of livelihood improvement and diversification to meet household wood demand and generating cash income.However,there is lack of information on the growth parmeters of Eucalyptus woodlot and the factors influencing the household decision on their establishment at the individual farmland level.The objective of this study was to examine local people’s knowledge on the adverse impacts and their attitudes towards growing Eucalyptus woodlot in Gudo Beret Kebele.We hypothesized that local people’s knowledge on the adverse impacts and their attitudes towards growing Eucalyptus woodlot in Gudo Beret Kebele is affected by socioeconomic and cognitive variables.Methods:A structured questionnaire comprising closed-and open-ended questions was developed and administered to a total of n=94 households to collect information on local people’s knowledge on the adverse impacts and their attitudes towards growing Eucalyptus woodlot.The households were randomly selected through a lottery system based on their house identification numbers.Descriptive statistics,binary logit,and multiple linear regression were used to analyze and interpret the data.Results:The results revealed that about 92%of the respondents noted that growing Eucalyptus woodlot had positive impacts on the socioeconomic situation of the community considering that it contributes to economic benefits through the sale of wood products,such as poles,construction materials,and fuelwood.However,only 8%of the respondents noted that the negative impacts of Eucalyptus woodlot were attributed to the decline in crop and forage production due to its allelopatic effect,and the reduction in ground water availability.Majority of the respondents(about 68%)preferred to grow Eucalyptus woodlot in Gudo Beret Kebele.Thus,most of the respondents(about 69%)had strongly agreed to have a positive attitude towards growing Eucalyptus woodlot.On the other hand,the binary logit regression model explained about 70.6%of the variance of local people’s knowledge on the adverse impacts of Eucalyptus woodlot.Overall,the multiple linear regression model revealed that socioeconomic and cognitive variables had significant effect on local people’s attitudes towards growing Eucalyptus woodlot(39.5%variance explained).Conclusions:We recommended that foresters,natural resource experts and managers,environmentalists,land use planners,and policy-makers should take the right and careful decision by assessing the overall socioeconomic and ecological aspects of Eucalyptus woodlot based on the interests of various stakeholders including local communities.

Regional differences of urbanization in China and its driving factors

After more than 30 years of rapid urbanization, the overall urbanization rate of China reached 56.1% in 2015.However, despite China's rapid increase in its overall rate of urbanization, clear regional differences can be observed. Furthermore, inadequate research has been devoted to in-depth exploration of the regional differences in China's urbanization from a national perspective, as well as the internal factors that drive these differences. Using prefecture-level administrative units in China as the main research subject, this study illustrates the regional differences in urbanization by categorizing the divisions into four types based on their urbanization ratio and speed(high level: low speed; high level: high speed; low level: high speed; and low level: low speed). Next, we selected seven economic and geographic indicators and applied an ordered logit model to explore the driving factors of the regional differences in urbanization. A multiple linear regression model was then adopted to analyze the different impacts of these driving factors on regions with different urbanization types. The results showed that the regional differences in urbanization were significantly correlated to per capita GDP, industry location quotients, urban-rural income ratio,and time distance to major centers. In addition, with each type of urbanization, these factors were found to have a different driving effect. Specifically, the driving effect of per capita GDP and industry location quotients presented a marginally decreasing trend, while main road density appeared to have a more significant impact on cities with lower urbanization rates.