In summary, this study advents six advanced ensemble linear regression-based prediction models: a novel ensemble linear regression-based prediction scheme for all possible periodic sub-sets of the given Time Series Set, a novel ensemble regression-based prediction scheme for all possible periodic sub-sets of the given Time Series Set, a model of a novel ensemble regression-based prediction scheme for all possible sub-sets of the given Time Series Set, a novel ensemble linear regression-based prediction scheme based on incorporation of the linear model strength contributed by each data point coordinate using a special weighted error metrics ensembling scheme that spans error metric types and considered recursive exhaustive linear regression models, and two novel modified linear regression-based prediction scheme and a model of Causal Cumulative logarithmic (recursive/non-recursive) exhaustive linear regression-based prediction scheme. A scheme for improving the theory of the novel all possible periodic sub-sets of the given Time Series set (non-recursive/ advanced recursive) ensemble linear regression scheme is also presented, along with a model of causal cumulative logarithmic (recursive/ non-recursive) exhaustive linear regression-based prediction scheme, analysis of the coefficient of relative data linearity of any given data of concern, and novel error minimization using forward and reversed prediction systems for the given series of concern.