Dynamics of Multicollinearity in Logistic and Probit Regression

Most often the social comparison or relative income hypothesis has been used as an explanation for the lack of systematic relationship between income and happiness, using the ordered probit regression method. The identification of relevant reference group and the estimation of the differential effects of comparison income have been controversial. To overcome these twin issues, this paper uses an ordinal comparison income approach based on rich/poor dichotomy and rank income. The rank income of an individual is defined as his relative position in the income distribution within the reference group and the average income of the reference group is used to define the rich/poor classification. The differential effects of ordinal incomes across life satisfaction distribution is estimated by the panel fixed effects ordered profit regression model using the WVS data for India. The estimated results show that ordinal income comparison, rather than cardinal average reference income, is a better predictor of life satisfaction levels. Raising income level is relatively important for less satisfied people while increasing rank status is important for highly satisfied people in India.

Download Full-text

Developmental milestones in early childhood and genetic liability to neurodevelopmental disorders

10.31234/osf.io/x8wst ◽

2020 ◽

Author(s):

Laurie John Hannigan ◽

Ragna Bugge Askeland ◽

Helga Ask ◽

Martin Tesli ◽

Elizabeth Corfield ◽

...

Keyword(s):

Language Development ◽

Motor Development ◽

Neurodevelopmental Disorders ◽

Autism Spectrum ◽

Probit Regression ◽

Developmental Milestones ◽

Genetic Liability ◽

Polygenic Scores ◽

Walking Age ◽

Children First

BackgroundEarly developmental milestones, such as the age at first walking or talking, are associated with later diagnoses of neurodevelopmental disorders, but the relationship to genetic risk for neurodevelopmental disorders are unknown. Here, we investigate associations between genetic liability to autism spectrum disorder (autism), attention deficit hyperactivity disorder (ADHD), and schizophrenia and attainment of early-life language and motor development milestones.MethodsWe use data from a genotyped sub-set (N = 15 205) of children in the Norwegian Mother, Father and Child Cohort Study (MoBa). In this sample, we calculate polygenic scores for autism; ADHD and schizophrenia and predict maternal reports of children’s age at first walking and talking, motor delays at 18 months, language delays at 3 years, and a generalized measure of concerns about development. We use linear and probit regression models in a multi-group framework to test for sex differences.ResultsADHD polygenic scores predicted earlier walking age in both males and females (β=-0.037, pFDR=0.001), and earlier first use of sentences (β=-0.087, pFDR=0.032) but delayed language development at 3 years in females only (β=0.194, pFDR=0.001). Additionally, we found evidence that autism polygenic scores were associated with later walking (β=0.027, pFDR=0.024) and motor delays at 18 months (β = 0.065, pFDR=0.028). Schizophrenia polygenic scores were associated with a measure of general concerns about development at 3 years in females only (β=0.132, pFDR=0.024).ConclusionsGenetic liabilities for neurodevelopmental disorders show some specific associations with measures of early motor and language development in the general population, including the age at which children first walk and talk. Associations are generally small and occasionally in unexpected directions. Sex differences are evident in some instances, but clear patterns across different polygenic scores and outcomes are hard to discern. These findings suggest that genetic susceptibility for neurodevelopmental disorders is manifested in the timing of developmental milestones in infancy.

Download Full-text

Explaining the informal economy in Western Europe: beyond a rational economic actor perspective

Journal of Economic Studies ◽

10.1108/jes-05-2020-0233 ◽

2020 ◽

Vol ahead-of-print (ahead-of-print) ◽

Author(s):

Colin Williams ◽

Gamze Oz-Yalaman

Keyword(s):

Informal Economy ◽

Western Europe ◽

Alternative Theory ◽

Probit Regression ◽

Content Type ◽

Social Actors ◽

The United Kingdom ◽

Economic Actor ◽

Expected Benefits ◽

Practical Implications

PurposeThe dominant theorisation of the informal economy views participants as rational economic actors operating in the informal economy when the expected benefits exceed the perceived costs of being caught and punished. Recently, an alternative theory has emerged which views participants as social actors operating in the informal economy due to their lack of vertical trust (in governments) and horizontal trust (in others). The aim of this paper is to evaluate these competing theorisations.Design/methodology/approachTo do so, data are reported from special Eurobarometer surveys conducted in 2007, 2013 and 2019 in eight West European countries (Austria, Belgium, France, Germany, Ireland, Luxembourg, the Netherlands and the United Kingdom).FindingsUsing probit regression analysis, the finding is that increasing the expected likelihood of being caught and level of punishment had a weak significant impact on the likelihood of participating in the informal economy in 2007, and there was no significant impact in 2013 and 2019. However, greater vertical and horizontal trust is significantly associated with a lower level of participation in the informal economy in all three time periods.Practical implicationsThe outcome is a call for a policy to shift away from increasing the expected level of punishment and likelihood of being caught, and towards improving vertical and horizontal trust. How this can be achieved is explored.Originality/valueEvidence is provided in a Western European context to support a shift away from a rational economic actor to a social actor approach when explaining and tackling the informal economy.

Download Full-text

Credit Constraint and Rural Household Welfare in the Mezam Division of the North-West Region of Cameroon

Sustainability ◽

10.3390/su13115964 ◽

2021 ◽

Vol 13 (11) ◽

pp. 5964

Author(s):

Louis Atamja ◽

Sungjoon Yoo

Keyword(s):

Regression Model ◽

Rural Household ◽

Household Welfare ◽

Switching Regression ◽

Probit Regression ◽

Credit Constraint ◽

Endogenous Switching Regression ◽

North West ◽

The North ◽

West Region

The purpose of this study is to examine the effect of the rural household’s head and household characteristics on credit accessibility. This study also seeks to investigate how credit constraint affects rural household welfare in the Mezam division of the North-West region of Cameroon. Using data from a household survey questionnaire, we found that 36.88% of the households were credit-constrained, while 63.13% were unconstrained. A probit regression model was used to examine the determinants of households’ credit access, while an endogenous switching regression model was used to analyze the impact of credit constraint on household welfare. The results from the probit regression model indicate the importance of the farmer’s or trader’s organization membership, occupation, and savings to the household’s likelihood of being credit-constrained. On the other hand, a prediction from the endogenous switching regression model confirms that households with access to credit have a better standard of welfare than a constrained household. From the results, it is necessary for the government to subsidize microfinance institutions, so that they can take on the risk of offering credit to rural households.

Download Full-text

The Impact of Climate Change Adaptation Strategies on Income and Food Security: Empirical Evidence from Small-Scale Fishers in Indonesia

Sustainability ◽

10.3390/su13147905 ◽

2021 ◽

Vol 13 (14) ◽

pp. 7905

Author(s):

Moh. Shadiqur Rahman ◽

Hery Toiba ◽

Wen-Chi Huang

Keyword(s):

Climate Change ◽

Food Security ◽

Climate Change Adaptation ◽

Substantial Effect ◽

Adaptation Strategies ◽

Small Scale ◽

Probit Regression ◽

The Government ◽

The Impact ◽

Climate Change Adaptation Strategies

The impacts of climate change on marine capture fisheries have been observed in several studies. It is likely to have a substantial effect on fishers’ income and food security. This study aims to estimate the impact of adaptation strategies on fishers’ income and their household’s food security. Data were collected from small-scale fishers’ households, which own a fishing boat smaller or equal to five gross tonnages (GT). The study sites were the two coastal regions of Malang and Probolinggo in East Java, Indonesia, due to the meager socioeconomic resources caused by climate change. A probit regression model was used to determine the factors influencing the fishers’ adaptation. Propensity score matching (PSM) was applied to evaluate the impact of the adaptation strategies on income and food security. Food security was measured by food consumption score (FCS). The findings indicated that participation in the fishers’ group affected adaptation strategies significantly, and so did the access to credit and climate information. Also, PSM showed that the adaptation strategies had a positive and significant impact on fishers’ income and food security. Those who applied the adaptation strategies had a higher income and FCS than those who did not. This finding implies that the fishery sector’s adaptation strategies can have significant expansion outcome and reduce exposure to risks posed by climate change. Therefore, the arrangement of more climate change adaptation strategies should be promoted by the government for small-scale fishers in Indonesia.

Download Full-text

Analysis of Single-Vehicle Crash Injury Severities in Urban River-Crossing Road Tunnels

Applied Mechanics and Materials ◽

10.4028/www.scientific.net/amm.743.526 ◽

2015 ◽

Vol 743 ◽

pp. 526-532 ◽

Cited By ~ 2

Author(s):

C.M. Jiang ◽

J.J. Lu ◽

L.J. Lu

Keyword(s):

Heavy Vehicles ◽

Probit Regression ◽

Factors Affecting ◽

Explanatory Variables ◽

Urban Road ◽

Road Tunnels ◽

Negative Effect ◽

Single Vehicle ◽

Injury Severities ◽

One Year

Based on the originally unprocessed data from the Official Platform of“110”Alarming Receiving Center (OP110ARC) of Shanghai Public Security Bureau (SPSB), 529 single-vehicle crashes reported during one year and a half which happened at the thirteen urban road tunnels going across the Huangpu River are used in this study. To investigate the factors affecting the crash influence severity levels, ordered probit regression is established. Several categories of factors are considered as explanatory variables in the models. The study finds that the entrance of the tunnels is the site where severe injury crashes trend to occur. Rainy and snowy days impose vehicles and motorists driving via the tunnel sections in danger. Tunnels with a low speed limit (40 km/h in this study) may be not as safe as we thought before. Two-wheel vehicles without sufficient physical protection for its drivers and heavy vehicles also show a negative effect on the operation safety of single-vehicle at these studied tunnels. Alcohol involved drivers are more likely to suffer from a severe crashes and gets badly hurt.

Download Full-text

Relationship between obesity and the risk of clinically significant depression: Mendelian randomisation study

The British Journal of Psychiatry ◽

10.1192/bjp.bp.113.130419 ◽

2014 ◽

Vol 205 (1) ◽

pp. 24-28 ◽

Cited By ~ 37

Author(s):

Chi-Fa Hung ◽

Margarita Rivera ◽

Nick Craddock ◽

Michael J. Owen ◽

Michael Gill ◽

...

Keyword(s):

Major Depression ◽

Causal Relationship ◽

Instrumental Variable ◽

Common Mental Disorders ◽

Genetic Risk Score ◽

Linear Regression Analysis ◽

Mendelian Randomisation ◽

Probit Regression ◽

Reverse Causality ◽

Increased Risk

BackgroundObesity has been shown to be associated with depression and it has been suggested that higher body mass index (BMI) increases the risk of depression and other common mental disorders. However, the causal relationship remains unclear and Mendelian randomisation, a form of instrumental variable analysis, has recently been employed to attempt to resolve this issue.AimsTo investigate whether higher BMI increases the risk of major depression.MethodTwo instrumental variable analyses were conducted to test the causal relationship between obesity and major depression in RADIANT, a large case–control study of major depression. We used a single nucleotide polymorphism (SNP) in FTO and a genetic risk score (GRS) based on 32 SNPs with well-established associations with BMI.ResultsLinear regression analysis, as expected, showed that individuals carrying more risk alleles of FTO or having higher score of GRS had a higher BMI. Probit regression suggested that higher BMI is associated with increased risk of major depression. However, our two instrumental variable analyses did not support a causal relationship between higher BMI and major depression (FTO genotype: coefficient −0.03, 95% CI −0.18 to 0.13, P = 0.73; GRS: coefficient −0.02, 95% CI −0.11 to 0.07, P = 0.62).ConclusionsOur instrumental variable analyses did not support a causal relationship between higher BMI and major depression. The positive associations of higher BMI with major depression in probit regression analyses might be explained by reverse causality and/or residual confounding.

Download Full-text

Rural–Urban Disparities in Access to Improved Sanitation in Indonesia: A Decomposition Approach

SAGE Open ◽

10.1177/21582440211029920 ◽

2021 ◽

Vol 11 (3) ◽

pp. 215824402110299

Author(s):

Sri Irianti ◽

Puguh Prasetyoputra

Keyword(s):

Rural Areas ◽

Urban Areas ◽

Universal Access ◽

Decomposition Analysis ◽

Multivariate Regression Analysis ◽

Probit Regression ◽

Decomposition Approach ◽

Explanatory Variables ◽

Improved Sanitation ◽

The Government

One of the targets in the Sustainable Development Goals (SDGs), which is Target 6.2, aims to achieve access to adequate and equitable sanitation. The Government of Indonesia targets universal access to improved sanitation in 2019. However, almost two out of five households in Indonesia are without access to improved sanitation. Moreover, access to improved sanitation is lower in rural areas than that in urban areas. Studies examining the drivers of the disparity in Indonesia are also limited. Therefore, this study was aimed at assessing the characteristics associated with the rural–urban disparity in access to improved sanitation facilities among households in Indonesia. We employed data from the 2016 Indonesian National Socio-Economic Survey (SUSENAS) comprising 290,848 households. The analysis was twofold. First, we fitted multivariate probit regression models using average marginal effects as the measure of association. We then conducted a detailed non-linear decomposition of the rural–urban disparity attributable to all the explanatory variables. The multivariate regression analysis suggested that households living in rural areas were 11.35% (95% confidence interval = [10.97, 11.72]) less likely to have access to improved sanitation facilities than those residing in urban areas. The decomposition analysis suggested that 48.78% are attributable to spatial, demographic, housing, and socio-economic factors, which meant that almost half of the inequalities could be reduced by equalizing these factors. The results provide a decomposition of factors amenable to curtail urban–rural inequalities. Hence, equity-oriented approaches to increasing access to improved sanitation should be prioritized to achieve universal access in 2030 in line with SDG Target 6.2.

Download Full-text

The Interaction between HIV and Intestinal Helminth Parasites Coinfection with Nutrition among Adults in KwaZulu-Natal, South Africa

BioMed Research International ◽

10.1155/2017/9059523 ◽

2017 ◽

Vol 2017 ◽

pp. 1-12 ◽

Cited By ~ 2

Author(s):

B. T. Mkhize ◽

M. Mabaso ◽

T. Mamba ◽

C. E. Napier ◽

Z. L. Mkhize-Kwitshana

Keyword(s):

South Africa ◽

Nutritional Status ◽

Transferrin Saturation ◽

Intestinal Helminth ◽

Helminth Parasites ◽

Probit Regression ◽

Cross Sectional ◽

Care Clinic ◽

Comprehensive Primary Health Care ◽

Kwazulu Natal

In South Africa few studies have examined the effects of the overlap of HIV and helminth infections on nutritional status. This cross-sectional study investigated the interaction between HIV and intestinal helminths coinfection with nutritional status among KwaZulu-Natal adults. Participants were recruited from a comprehensive primary health care clinic and stratified based on their HIV, stool parasitology, IgE, and IgG4 results into four groups: the uninfected, HIV infected, helminth infected, and HIV-helminth coinfected groups. The nutritional status was assessed using body mass index, 24-hour food recall, micro-, and macronutrient biochemical markers. Univariate and multivariate multinomial probit regression models were used to assess nutritional factors associated with singly and dually infected groups using the uninfected group as a reference category. Biochemically, the HIV-helminth coinfected group was associated with a significantly higher total protein, higher percentage of transferrin saturation, and significantly lower ferritin. There was no significant association between single or dual infections with HIV and helminths with micro- and macronutrient deficiency; however general obesity and low micronutrient intake patterns, which may indicate a general predisposition to micronutrient and protein-energy deficiency, were observed and may need further investigations.

Download Full-text

State failure, genocide and politicide reconsidered

The British Journal of Politics and International Relations ◽

10.1177/13691481211044654 ◽

2021 ◽

pp. 136914812110446

Author(s):

Gary Uzonyi

Keyword(s):

Government Policy ◽

Strong Support ◽

State Failure ◽

Probit Regression ◽

Bivariate Probit ◽

Formal Institutions ◽

Physical Ability ◽

Government Institutions ◽

Lower Likelihood

State failure is highly threatening to incumbent elites, providing them motive to engage in genocide or politicide. The destruction of formal institutions may also remove political barriers to employing such atrocities. Thus, previous scholars expect state failure to increase the likelihood of genocide and politicide. However, such logic overlooks that these atrocities are government policy, which require planning and implementation. Since government institutions collapse during state failure, regime elites will possess little physical ability to commit widespread and systematic killing of their enemies. Therefore, I expect state failure to be associated with a lower likelihood of genocide and politicide. Using bivariate probit regression to handle the endogeneity between state failure and these atrocities, I find strong support for my argument in all country-years since 1946.

Download Full-text