The socioeconomic determinants of body mass index: A conditional quantile regression analysis

The presence of good and healthy nutrition is linked to the population of healthy individuals in a country. The Body Mass Index (BMI) serves as a valuable tool for assessing a person's nutritional status, determining if they are underweight, overweight, obese, or at a healthy weight relative to their height. Nevertheless, BMI should not be interpreted as a direct indicator of body fat percentage. When an individual's BMI falls outside the healthy BMI range, their susceptibility to disease or mortality may be considerably heightened. BMI for women is influenced by factors such as age and living conditions and is commonly known as BMI for age. Excessive body fat in women can contribute to weight-related illnesses and other health complications, while being underweight can also pose health risks. In statistical and econometric studies, it is critical that the functional structure of the model meets the model's basic assumptions. The least squares (LSC) method, which is one of the parametric methods that examines the relationship between variables, has some basic assumptions, such as the normality assumption that the data does not have extreme values and the equal variance assumption. When these assumptions are not met, using these models causes various problems. The quantile regression method does not require any distribution assumptions, and since it estimates parameter coefficients according to various quantiles, it gives better estimates than traditional methods in data sets with extreme values. This study aims to examine the factors affecting women's BMI using the quantile regression (QR) technique developed by Koenker and Basset (1978). Quantile regression results showed that socio-economic variables such as age, income, television viewing habit, number of child bearings, and education level had a significant impact on women's BMI.