Yildiz Technical University, Istanbul, Turkey.
Nonlinear Dynamics Psychol Life Sci. 2021 Apr;25(2):207-236.
The chaotic structure of air pollution, human health, the evidence of cointegration and causality between the air pollution, transportation, tourism, construction, and manufacturing industries and the evidence of cointegration and causality between the air pollution, transportation, tourism, construction, and manufacturing industries and mortality rate were explored for China, India, and Turkey in the period of 1975-2018 by using four different methods. Firstly, Lyapunov and Shannon tests were applied to determine chaotic dynamics. The maximum Lyapunov test, for the selected variables, found the evidence of chaotic dynamics. Secondly, Fourier ADF and NL tests were applied. Fourier unit root test determined stationary of the variables. Thirdly, bootstrapping autoregressive distributed lag with Fourier transformation (FBARDL) was used to determine the evidence of cointegration between the variables with two different models. FBARDL test determined that the manufacturing, tourism, transport, and construction industries, air pollution, and mortality rate have evidence of cointegration in different two models. Lastly, the causality test with Fourier transformation was used to determine the direction of causality between the variables. Granger causality test determined that there is evidence of one-way causality running from transportation, tourism, construction, and manufacturing industries to air pollution and mortality rates. Accordingly, the results of this paper suggest air pollution and human health have chaotic behaviors. Air pollution has a complex, multi-variable, and multi-coupling system. Air pollutants influencing factors and air pollution itself have adverse effects on human health.
采用四种不同方法，探索了 1975-2018 年中国、印度和土耳其的空气污染、人类健康、空气污染与交通、旅游、建筑和制造业之间的协整和因果关系的混乱结构，以及空气污染与交通、旅游、建筑和制造业之间的协整和因果关系的证据。首先，应用 Lyapunov 和 Shannon 检验来确定混沌动力学。对于所选变量，最大 Lyapunov 检验发现了混沌动力学的证据。其次，应用 Fourier ADF 和 NL 检验。Fourier 单位根检验确定了变量的平稳性。第三，采用带 Fourier 变换的自回归分布滞后 Bootstrap（FBARDL）来确定变量之间的协整证据，使用了两种不同的模型。FBARDL 检验确定了制造业、旅游业、运输业和建筑业、空气污染和死亡率在两个不同模型中有协整的证据。最后，采用 Fourier 变换的因果关系检验来确定变量之间的因果关系方向。格兰杰因果关系检验确定了从交通、旅游、建筑和制造业到空气污染和死亡率存在单向因果关系的证据。因此，本文的结果表明，空气污染和人类健康具有混沌行为。空气污染具有复杂的多变量和多耦合系统。空气污染的影响因素和空气污染本身对人类健康有不良影响。
