Applying bootstrap quantile regression for the construction of a low birth weight model

Abstract

Background: Most investigators use ordinary least squares (OLS) methods to model low birth weight. When the data are non-normal or contain outliers, OLS become ineffective. However, the quantile method of forecasting low birth weight has not been fully evaluated, although it has good potential for overcoming problems associated with linear regression. Methods: The present study reports our comparison between the OLS and quantile regression methods for modeling low birth weight when the data are right skewed and outliers are presented. Additionally, we evaluated the performance of the associated algorithm in recovering the true parameter using the bootstrap method. Results: Our study found that a mother's education level, the number of maternal parities, and the last birth interval significantly impacted low birth weight at any selected low quantile. Based on the bootstrap simulation study, the proposed model was considered to be acceptable since both methods generated nearly identical estimates of the parameter model. An accuracy test proved that the quantile method was an unbiased estimator. Conclusions: The present study found that low birth weight is significantly affected by the mother's educational level, the number of maternal parities, and the last birth interval.

Bayesian LASSO Quantile Regression: An Application to the Modeling of Low Birth Weight

“…The empirical data which is used in this study is right skewed or not normal distribution. The appropriate method to model such this data is quantile regression approach [3] [12]. We then combine the Bayesian approach and quantile regression since the size of data is moderate, 150 sample only.…”

“…Meanwhile categorical types are residence, education level, and sex of the baby [2]. Descriptives of the data can be seen in Yanuar et al [12]and not presented here because of limited space.…”

“…The secondary data from study of Yanuar et al is used in this research [12]. The size of sample data is 150 respondents of mothers who just delivered a singleton live birth and were living in West Sumatera, Indonesia.…”

“…There are not much work has been developed under the Bayesian framework especially for modeling the determinants of low birth weight. Therefore, this study will combine the quantile regression with Bayesian LASSO approach to model the low birth weight [11].Secondary data of birth weight in Sumatera Barat at 2016 from Yanuar et al [12] will be used. Thus, this study is also important to be done since there are no previously published manuscripts related to model the birthweight in West Sumaterausing Bayesian quantile regression and Bayesian LASSO quantile method.…”

Determining of Risk Factors of Low-Birthweight Babies in Padang, West Sumatra Using Logistic Regression Analysis

“…From a search conducted on several research results, there are factors that influence the occurrence of LBW including pregnant women at the age of less than 20 years or more than 35 years, the distance of pregnancy is too short (less than 2 years), mothers with previous LBW conditions, doing physical work several hours without rest, socioeconomic, poor nutritional status and smokers, drug users and alcohol. Other factors which can also be suspected to influence are parity, the amount of weight gain during pregnancy, prenatal care and haemoglobin concentration [7], [8], [9], [10], [11]. In addition, several studies have shown that there is a fairly close relationship between the incidence of LBW and premature birth [12].…”