Nonlinear Decomposition Technique Research Articles

The Role of Public Assistance in Improving Food Security During Covid-19 in Togo

This paper provides insights into the potential implication of public assistance on food insecurity in the Togolese population during the Covid-19 period. The Oaxaca–Blinder Non-Linear decomposition technique and binary logit model were applied to Data of Round 8 of Afrobarometer of Togo. The empirical results reveal that women were more food insecure in Togo during the Covid-19 period. This particular vulnerability of women was explained by the difference in economic resources, place of residence, and response measures put in place by the State. However, the public assistance received has helped to reduce the gaps observed in food insecurity. The study reveals the importance of public assistance measures, especially those based on gender, during shocks that may occur in an economy.

PS-P01-7: THE COHERENCE OF INFECTIOUS AND CHRONIC DISEASE EPIDEMICS; AMONG ELDERLY IN RURAL URBAN PERSPECTIVES

Background: Chronic diseases are the leading causes of disability and premature death among the elderly population in Uganda. The study, using data from the National Statistical office through censuses and surveys, examined the prevalence and determinants of chronic diseases among the population aged 60 + in Uganda by applying bivariate and logistic regression analyses and used a non-linear decomposition technique to understand the urban-rural differences in the coherence of chronic diseases. Methods: About 25% of the elderly in Uganda reportedly have at least one chronic disease. Seventeen percent elderly in rural areas and 29% in urban areas suffer from a chronic disease. Hypertension and diabetes account for about 68% of all chronic diseases. Those with higher levels of education, staying in urban areas, those who are economically dependent on others, staying alone or without spouse and children, and belonging to wealthy households have a higher likelihood of having a chronic disease. Results: The probability of having a chronic disease is 1.15 times higher among urban residents as compared to their rural counterparts. Elderly rural women, compared to elderly rural men, and never-married, widowed, and divorced elderly urban women, compared to married elderly urban men, are significantly more likely to suffer from chronic ailments. Differences in education, wealth status, and caste are the three most significant contributors to the urban-rural gap in chronic diseases. Conclusion: The high risk of chronic diseases among certain subsets of the elderly population must be recognized as a key public health concern. The findings of our study will likely help promote healthy ageing in Uganda.

Decomposing the Intraurban Malnutrition Gap Between Poor and Non-poor Children in Colombia : Decomposing the Intraurban Malnutrition Gap Between Poor and Non-poor Children in Colombia.

In Colombia, although it can be said that, on average children living in urban areas have better quality of life than their rural peers, it is also true that within cities, there are high levels of socioeconomic inequality. Our objective is to identify the contribution of the factors that explain the gap in stunting and excess weight between poor and non-poor children under 5years of age in urban areas of Colombia. We use data from the 2015 National Nutritional Status Survey, and two nonlinear decomposition techniques based on the classical decomposition method developed by Blinder-Oaxaca. With a sample of 6877 observations, the results show that the intraurban gap of stunting between poor and non-poor children in urban areas is 4.8 percentage points. Its main determinants are the mother's educational level (46.5%), affiliation to the health system by the mother (19.4%), and assisted delivery in a medical institution (16.6%). For excess weight, the gap is - 2.1 percentage points, and its main determinants are the mother's educational level (39.2%) and birth attended by a physician (21.8%). This study suggests the coexistence of a double burden of malnutrition (DBM) in children under 5years of age living in urban areas of Colombia. Stunting is associated with low-income levels while excess weight is associated with higher income levels. The identification of the main determinants of DBM and its relative importance, constitutes a contribution for public policy makers aimed at reducing socioeconomic gaps.

Economic inequalities and rural-urban disparities in clean cooking fuel use in Ghana

Expanding access to clean cooking fuels is an important driver of sustainable development goal (SDG) 7, which is to ensure universal access to affordable and clean energy. Ghana is among 20 countries that account for more than 80% of the global deficit in access to clean cooking fuel, despite efforts to improve uptake of clean cooking fuel over the years. Using the last three rounds of the Ghana Living Standards Survey, this study examines the trend and extent of economic inequalities, as well as the factors that account for the rural-urban disparity in the use of clean cooking fuels. We first used concentration curves and concentration indices to analyse the magnitude and trend of economic inequality. We then decomposed the rural-urban gap in the use of clean cooking fuels by applying a non-linear decomposition technique. Our results reveal that, in general, economic inequality persists across the three survey rounds. The level of inequality was higher in urban areas but relative change over time was higher in the rural sample. Also, rural-urban differences in economic status, education of the household head and ownership of television were the biggest contributors to the rural-urban gap in clean cooking fuel use. Our findings reveal the need for a more targeted approach in the design and implementation of policies to encourage the use of clean cooking fuels. Tailored strategies to reach the poor and vulnerable in rural and urban areas while scaling up access will be a step in the right direction.

Prevalence and potential determinants of chronic disease among elderly in India: Rural-urban perspectives.

Chronic diseases are the leading causes of disability and premature death among the elderly population in India. The study, using data from the 75th round of the NSSO survey (N = 44,631), examined the prevalence and determinants of chronic diseases among the population aged 60+ in India by applying bivariate and logistic regression analyses and used a non-linear decomposition technique to understand the urban-rural differences in the prevalence of chronic diseases. About 21% of the elderly in India reportedly have at least one chronic disease. Seventeen percent elderly in rural areas and 29% in urban areas suffer from a chronic disease. Hypertension and diabetes account for about 68% of all chronic diseases. The prevalence of chronic diseases is the highest in Kerala (54%), followed by Andhra Pradesh (43), West Bengal (36), and Goa (32). Those with higher levels of education, staying in urban areas, those who are economically dependent on others, staying alone or without spouse and children, and belonging to wealthy households have a higher likelihood of having a chronic disease. The probability of having a chronic disease is 1.15 times higher among urban residents as compared to their rural counterparts. Elderly rural women, compared to elderly rural men, and never-married, widowed, and divorced elderly urban women, compared to married elderly urban men, are significantly more likely to suffer from chronic ailments. Differences in education, wealth status, and caste are the three most significant contributors to the urban-rural gap in chronic diseases. The high risk of chronic diseases among certain subsets of the elderly population must be recognized as a key public health concern. The findings of our study will likely help promote healthy ageing in India.

Improving GPR Imaging of the Buried Water Utility Infrastructure by Integrating the Multidimensional Nonlinear Data Decomposition Technique into the Edge Detection

Although ground-penetrating radar (GPR) is effective to detect shallow-buried objects, it still needs more effort for the application to investigate a buried water utility infrastructure. Edge detection is a well-known image processing technique that may improve the resolution of GPR images. In this study, we briefly review the theory of edge detection and discuss several popular edge detectors as examples, and then apply an enhanced edge detecting method to GPR data processing. This method integrates the multidimensional ensemble empirical mode decomposition (MDEEMD) algorithm into standard edge detecting filters. MDEEMD is implemented mainly for data reconstruction to increase the signal-to-noise ratio before edge detecting. A quantitative marginal spectrum analysis is employed to support the data reconstruction and facilitate the final data interpretation. The results of the numerical model study followed by a field example suggest that the MDEEMD edge detector is a competent method for processing and interpreting GPR data of a buried hot spring well, which cannot be efficiently handled by conventional techniques. Moreover, the proposed method should be readily considered a vital tool for processing other kinds of buried water utility infrastructures.

Crisis, Adjustment and Resilience in the Greek Labor Market: An Unemployment Decomposition Approach

The crisis in Greece led to one of the largest economic shocks in European history. Drawing on micro-data from the Greek Labour Force Survey, we utilize standard micro-econometric methods and non-linear decomposition techniques to measure the size of the shock exerted on the Greek regional and national labor markets and the compositional and price adjustments in response to this. We find elements of economic dynamism, with some sizeable price adjustments in the economy of the Greek capital, Athens; but overall our results show that compositional adjustments (in labor quality/characteristics) have been partial and limited, becoming stronger only in the more recent recovery. Our results suggest a significant metropolitan advantage with regard to economic resilience, coming predominantly from a more efficient functioning of the labor market in metropolitan areas vis-a-vis other regions. Our use of the decomposition techniques for the analysis of macro-level developments in the labor market offers a novel perspective to the application of the decomposition methodology.

Caste/ethnic differences in school dropout among 5-20-year olds in Nepal

ABSTRACT Reducing school dropout rates has been a challenging policy problem in an effort to improve educational participation and attainment in developing countries. This paper examines the differences in school dropout rates among the 5–20-year olds across three major caste/ethnic groups in Nepal using nationally representative survey data. Findings suggest sizeable differences witnessed by the indigenous and especially lower-caste Dalit groups when compared to the historically privileged Hindu groups. The use of Fairlie’s non-linear regression-based decomposition technique helps ascertain the major sources of these differences. Parental occupations, age, marital status and mountains, hills and far-western regions of residence are found to help explain such differences. The paper underscores the role of public policies on affirmative action and area-based, conditional cash transfer initiatives as well as effective public awareness campaigns to improve the educational participation of these otherwise historically disadvantaged caste/ethnic groups in Nepal.

Guideline-Concordant Treatment Among Elderly Women With HER2-Positive Metastatic Breast Cancer in the United States.

It is crucial to identify whether women with HER2-positive (HER2+) metastatic breast cancer (MBC) are treated according to treatment guidelines and whether treatment disparities exist. This study examined guideline-concordant treatment among women with HER2+ MBC and determined the magnitude of differences in treatment between those with positive and negative hormone receptor (HR) status using a nonlinear decomposition technique. A retrospective observational cohort study was conducted using the SEER-Medicare linked database. The study cohort consisted of women aged ≥66 years diagnosed with HER2+ MBC in 2010 through 2013 (n=241). Guideline-concordant initial treatment after cancer diagnosis was defined based on the NCCN Clinical Practice Guidelines in Oncology for Breast Cancer. A multivariable logistic regression was performed to identify significant predictors of guideline-concordant treatment. A postregression decomposition was conducted to identify the magnitude of disparities in treatment by HR status. Of 241 women included in the study, a total of 76.8% received guideline-concordant treatment. These women were significantly more likely to have positive HR status (P=.0298), have good performance status (P=.0009), and more oncology visits (P<.0001). With 1-year increments in age at cancer diagnosis, the likelihood of receiving guideline-concordant treatment reduced by 5% (P=.0356). The decomposition analysis revealed that 19.0% of the disparity in guideline-concordant treatment between women with positive and negative HR status was explained by differences in their characteristics. Enabling characteristics (marital status, income, and education) explained the highest (22.8%) proportion of the disparity. Nearly one-quarter of the study cohort did not receive guideline-concordant treatment. Our findings suggest opportunities to improve cancer care for elderly women with negative HR status who are unpartnered or have lower socioeconomic status. The high unexplained portion of the disparity by HR status can be due to patient treatment preferences, propensity to seek care, and organizational and physician-level characteristics that were not included in the study.

Age at first marriage, age at first sex, family size preferences, contraception and change in fertility among women in Uganda: analysis of the 2006\u20132016 period

BackgroundUganda’s fertility was almost unchanging until the year 2006 when some reductions became visible. Compared to age at first marriage and contraceptive use, age at sexual debut and family size preferences are rarely examined in studies of fertility decline. In this study, we analyzed the contribution of age at first marriage, age at first sex, family size preferences and contraceptive use to change in fertility in Uganda between 2006 and 2016.MethodsUsing data from the 2006 and 2016 Uganda Demographic and Health Survey (UDHS), we applied a nonlinear multivariate decomposition technique to quantify the contribution of age at first marriage, age at first sex, family size preference and contraceptive use to the change in fertility observed during the 2006–2016 period.ResultsThe findings indicate that 37 and 63% of the change in fertility observed between 2006 and 2016 was respectively associated with changing characteristics and changing fertility behavior of the women. Changes in proportion of women by; age at first marriage, age at first sex, family size preferences and contraceptive use were respectively associated with 20.6, 10.5 and 8.4% and 8.2% of the change in fertility but only fertility behavior resulting from age at first sex was significantly related to the change in fertility with a contribution of 43.5%.ConclusionsThe study quantified the contribution of age at first marriage, age at first sex, family size preferences and contraceptive use to the change in fertility observed between 2006 and 2016. We highlight that of the four factors, only age at sexual debut made a significant contribution on the two components of the decomposition. There is need to address the low age at first sex, accessibility, demand for family planning services and youth-friendly family planning services to young unmarried women such that they can achieve their desired fertility. The contribution of other factors such as education attainment by women and place of residence and their relationship with changes in fertility calls for addressing if further reduction in fertility is to be realised.

Race/ethnic Differences in Spatial Distance between Adult Children and Their Mothers.

This brief study examines race/ethnic differences in geographic distance to mothers among adults in the U.S. Race/ethnic differences in rates of adult children living with their mothers in the U.S. are well documented but spatial distances beyond shared housing are not. Spatial distances between residential locations of adults 25 and older and their biological mothers are estimated using the 2013 Panel Study of Income Dynamics for Hispanics, blacks, and whites. Multinomial logistic regression models and nonlinear decomposition techniques are used to assess the role of demographic factors, socioeconomic status, and health of the child and mother in accounting for race/ethnic differences in adult child-mother proximity. Blacks are more likely than whites to live with their mother and more likely to live within 30 miles but not coresident, whereas whites are more likely to live more than 500 miles away. Geographic proximity to mother is distinct for Hispanics with nearly one-third having their mother outside the U.S. Demographic, socioeconomic, and health factors account for the fact that blacks are about twice as likely as whites to live with their mother but do not fully account for large white-black differences in proximity outside the household. The single most important factor accounting for white-black differences is marital status for coresidence but education for proximity in the U.S. beyond coresidence. New national estimates illustrate the complexity of race/ethnic differences in proximity to mothers that are not reflected in studies of coresidence.

Adaptive learning-based finite-time performance of nonlinear switched systems with quantization behaviors and unmodeled dynamics

This paper proposes a novel finite-time adaptive learning strategy for a class of nonlinear switched systems with quantization behaviors and unmodeled dynamics under unpredictable switchings. The neural network learning framework is introduced to manifest the characteristics of nonlinear systems to obtain better performances. The system nonlinearities include more complicated non-strict feedback structure and full-state dependent unmodeled dynamics. In virtue of the nonlinear decomposition technique and the finite-time criterion, an adaptive learning strategy is developed step by step. Under the proposed strategy, a common Lyapunov function for all the subsystems is constructed to guarantee that in finite time all the signals of the switched system converge to a small domain near the origin under arbitrary switchings. An illustrative example is finally given to verify the effectiveness of the proposed main results.

HSR19-111: Disparities in Guideline-Concordant Care Among HER2 Positive Metastatic Breast Cancer Patients

Background: We examined guideline-concordant care among women with HER2+ MBC and determined the magnitude of differences in guideline-concordant care between those with positive and negative hormone receptor (HR) status by utilizing a non-linear decomposition technique. Methods: We conducted a retrospective observational cohort study using the Surveillance, Epidemiology, End Results-Medicare linked database. The study cohort consisted of women age &gt;66 years diagnosed with HER2+ MBC in 2010–2013 (N=241). Guideline-concordant initial care within 6 months of cancer diagnosis was defined as per NCCN Clinical Practice Guidelines in Oncology (NCCN Guidelines). A multivariate logistic regression was performed to identify the significant predictors of guideline-concordant care. A post-regression non-linear decomposition was conducted to examine the magnitude of disparities in guideline concordant care by women’s HR status. Results: 76.8% of the study cohort received guideline-concordant care, while 23.2% did not. As compared to those who did not receive guideline-concordant care, women who received guideline-concordant care were significantly more likely to have positive HR status (adjusted odds ratio (AOR)=2.11; P=.04), had good performance status (AOR=3.46; P=.0008), and had a higher number of oncology visits (AOR=8.05; P&lt;.0001). With 1 year increase in age at cancer diagnosis, there was 5% lesser likelihood of receiving guideline-concordant care (AOR=0.95; P=.04). From the decomposition analysis, 19.0% of the disparity in guideline-concordant care between women with positive and negative HR status was explained by differences in their characteristics. Enabling characteristics (marital status, census-level income, and education) explained the highest (22.8%) proportion of the disparity, followed by external environmental factors (location of residence, SEER region, hospitals offering oncology services) at 5.3%, and need-related factors (tumor grade, comorbidity, performance status, number of metastases) at 3.2%. Conclusion: Almost one quarter of the study cohort did not receive guideline-concordant care. There are opportunities to improve cancer care for women with negative HR status who have lower socioeconomic status. The high unexplained portion of differences in guideline-concordant care (81.0%) can be due to patient preferences for treatment, propensity to seek care, and organizational and physician-level factors not captured in the database.

GENDER DIFFERENCES IN ATTITUDES TOWARD RISK: EVIDENCE FROM ENTREPRENUERS IN GHANA AND UGANDA

The literature on risk aversion suggests that women are less likely to be risk loving than men in financial and insurance decision-making by virtue of their sex and biological make-up. This paper tests this assertion using a unique dataset collected in Ghana and Uganda and assesses the gender differences in self-reported risk perceptions of entrepreneurs by applying a non-linear decomposition technique. The results indicate that on average, entrepreneurs in Ghana report to be less risk loving their counterparts in Uganda. Furthermore, female entrepreneurs are less likely to report to be risk loving compared to male entrepreneurs in both countries. The results from the decomposition analysis show that gender differences in risk perceptions arise mainly from the unexplained component. For Ghana in particular, the findings show that the gender differences in self-reported risk perceptions stems from differences in education and previous business experience.

Gender and Entrepreneurship: Empirical Analysis Using the German Socio-Economic Panel (SOEP)

This study investigates the gender disparity in entrepreneurship in Germany. Using the data from the Socio-Economic Panel (SOEP), levels of education, Risk tolerance, and parental role model have a positive significant influence on the probability of Women’s Self-employment. I find no evidence that the higher educational level plays a role in Men’s probability of self-employment. Also, there is a higher probability of self-employment activity in East Germany than in West Germany. In West Germany, strong evidence prevails that both self-employed father and mother has a positive influence on Women’s probability of self-employed but not for men. Yet, in East Germany, only self-employed mother's has a significant influence on women. Non-linear Blinder-Oaxaca decomposition technique revealed that Women have a less probability of self-employment than Men. On average more than one-third of the gender difference driven from risk aversion behavior of women.

Update on Racial Disparities in Access to Healthcare: An Application of Nonlinear Decomposition Techniques*

ObjectiveAccess to care disparities between non‐Hispanic whites and Hispanics continue to persist; updating the literature is essential to finding solutions.MethodsUsed the Medical Expenditure Panel Survey 2009–2011, Area Health Resource File, and 2010 Census to update the literature using nonlinear decomposition techniques.ResultsFindings confirm previous studies: health insurance is essential in reducing racial disparities in healthcare. The impact of language has grown to 25 percent. Findings also suggest that acquiring public health insurance and gaining citizenship offset the access to care disparities between non‐Hispanic whites and Hispanics, especially for women. Heterogeneity exists between Hispanic women and men, some of which comes from differences between their attitudes and beliefs regarding healthcare. However, these disparities cannot be explained with observed characteristics.ConclusionDisparities in access to care have changed and observed characteristics explain all disparities between non‐Hispanic whites and Hispanics. Hispanic men suffer disproportionately from access to care disparities compared to Hispanic women.

The evolution of female labour force participation in urban China

AbstractStarting from the late 1980s, despite rapid economic growth, female labour force participation in urban China has shown a general declining trend. Using repeated cross‐sections from the Chinese Household Income Project Series (CHIP), this paper attempts to systematically relate the decrease in the labour force participation of married women to the socio‐economic changes happening in urban China during the same period of time. Adopting both linear and nonlinear decomposition techniques, the results indicate that the changes in married women's labour force participation during the periods 1995–2002 and 2007–2013 can be explained by the concurrent changes in the distribution of socio‐economic variables, while the changes during the periods 1988–1995 and 2002–2007 are mostly driven by the leftward shift of married women's participation function.

Detection of Third Heart Sound Using Variational Mode Decomposition

In this paper, a novel approach for the separation of heart and lung sounds (HLS) is proposed based on the nonlinear decomposition technique, variational mode decomposition (VMD) of phonocardiogram signal. Subjective validation and objective quantification are used to assess the performance of the method. The proposed technique is found to perform better than the conventional algorithms such as empirical mode decomposition (EMD), ensemble EMD, and different variants of the EMD. As the third heart sound, $S_{3}$ is an important clinical sign of cardiac failure in elderly patients, a new technique is proposed for its detection. This method is built on the VMD and the smoothed pseudo Wigner–Ville distribution. In total, 40 sets of cardiac cycles containing $S_{3}$ are obtained from publicly available databases and are used to evaluate the performance of the proposed method in noiseless condition. It is able to detect the $S_{3}$ correctly even when the normalized amplitude of $S_{3}$ is 14.1%, whereas the existing method based on the Hilbert variation decomposition requires at least 16.13% of the normalized amplitude of $S_{3}$ in comparison to the normalized amplitude of the highest peak present in the cardiac cycle. In addition to this, the result shows that the proposed method can detect $S_{3}$ in noisy cases up to SNR level of −5 dB.

Decomposing the gap in child malnutrition between poor and non-poor in Sierra Leone

The main aims of the study were to quantify the contribution of factors that explain the poor/nonpoor gap in underweight and stunting among children aged less than 5 years in Sierra Leone. We used cross-sectional data from the Sierra Leone DHS conducted during 2013. Descriptive statistics, logistic regression, and nonlinear (fairle) decomposition technique were used to explain the contribution to the average gap in malnutrition between poor and non-poor children in Sierra Leone. Considerable proportions of children were found to be underweight (16.8%) and stunted (38%) in Sierra Leone in 2013. The malnutrition gap between the poor and non-poor was stark in Sierra Leone. For these two indicators, the main contributing factors were place of residence, mother’s education, media exposure, and institutional delivery. Based on the findings, the study suggests that improving public services such as basic health care and the education level of the mothers among the poor can ameliorate the negative impact of poverty on childhood malnutrition.

Factors explaining the North–South differentials in contraceptive use in Nigeria: A nonlinear decomposition analysis

Background: Northern Nigeria has some of the worst reproductive health indicators worldwide. Conspicuous North–South variations exist in contraceptive use; not much is known about the drivers of contraceptive use disparities in the North compared to the South. Objective: In this study, we examine the relative weights of the factors that contribute to this North–South gap in contraceptive prevalence. Methods: Using the women’s 2013 Demographic Health Survey dataset, we applied a nonlinear decomposition technique to determine the contribution of sociodemographic and socioeconomic characteristics, conjugal relationship dynamics, intimate partner violence, ideational variables, and Islamic culture to the North–South disparities in contraceptive use. Results: There was a gap of 12.4 percentage points in contraceptive prevalence between the north and south of Nigeria (5.2% vs 17.6%). The largest contributors to the gap were ideational characteristics (explaining 42.0% of the gap) and socio-economic profiles (explaining 42.6%). Patterns of conjugal relationship dynamics (11.1%), socio-demographic characteristics (‒11.0%), Islamic religious culture (7.6%), and exposure to family planning messaging (6.1%) were also significant contributors. Conclusions: Effective interventions to increase contraceptive use in northern Nigeria should aim at addressing socioeconomic disadvantage in the North, impacting ideational characteristics and specifically targeting poor women and those with low levels of education. Working with Islamic religious leaders is also critical to bridging the gap. Contribution: This paper broadens the knowledge on the determinants of contraceptive use in Nigeria by identifying contextual factors that operate differently in the North compared to the South.