|
To Use Quantile Mixed Regression for Analyzing Iranian Households Income and Expenditure
|
Enayat Barani   , Mousa Golalizadeh * |
| Tarbiat Modares University |
|
|
Abstract: (3074 Views) |
| One of the essential assumptions while working with the linear regression models based on the variation of the mean of response variables is independency among error components. To fail this assumption guide researchers to utilize linear mixed effect models. But, to use the latter models is dependent on absence of outliers, one can employ the linear mixed quantile models where error component follows asymmetric Laplace distribution. In this paper, the performance of the available frequentist estimation methods in these models was first evaluated through simulation studies. It is seen that the combined methods approximated quadrature Gauss and unsmooth optimization algorithm leads to more accurate estimators in compare with the stochastic approximation of expectation maximization algorithm. Then, due to dependency between the expenditure as well as having outliers among households in the provinces, the model has been used to analyze the Iranian households’ income and expenditure data in urban area which is collected at 2010. Regarding the results gained, it is expected that households with more income tend to pay more on their costs. On the other hand, the area of housing unit, the population of the province, number of employed members of household and dewelling occupation have less effect on the gross cost of households. In addition, employement statue of the older head of household leads to a reduction in the gross cost. |
|
| Keywords: Asymmetric Laplace distribution, Linear quantile mixed regression models, Gaussian quadrature approximations method, expectation maximization algorithm, income and expenditure data. |
|
|
Full-Text [PDF 256 kb]
(2368 Downloads)
|
Type of Study: Applicable |
Subject:
Special
Received: 2019/06/2 | Accepted: 2019/12/8 | Published: 2019/12/18
|
|
|
|
|
|
|
| Send email to the article author |
|
|
