Abstract

    Evidences indicated that the effects of socioeconomic cycle on health outcome in population-base studies tend to lead to erroneous conclusions if the time effects problems are neglected. Transfer function model is considered can improve the output variable time series model by analyzing the dynamic relation between input time series data and output time-series data. Nonetheless, ecological problem need to be concerned. While the individual-base study may avoid the faculty, the collinear or interaction among multifactor raises the difficulty of analysis and explanation. Additionally, because of strong correlations between risk factors and within risk factor across the life course, teasing apart the effects of early versus later life exposure to each individual risk factor is difficult. The life course and health perspective considers chronic disease in terms of the social and physical hazards, and the consequent biological, behavioral and psychosocial processes, that operate across stages of the life span to cause or modify risk of disease.