Many solar activities, e.g. sunspots, Mg II indices, coronal index of solar activity, etc., are produced by solar magnetic fields that affect both, the solar atmosphere and heliosphere, including the terrestrial environment. The atmosphere of the sun is recognized by photosphere, as a visible solar disk and layers above it: chromosphere, transition zone and corona. The nature of solar activity can be, in the emission solar corona, recognized by the coronal index of solar activity (CI). Its dataset in the period 1944–2009 will be used to test the ARIMA model. Based on the minimum AIC and BIC values, ARIMA(1,0,1) model is selected for each CI cycle and total dataset. The quality of the selected model is endorsed by (RMSE, MAPE, & FPE). The behaviour of the observed (original) CI cycle with predictive CI cycle shows a strong correlation, of more than 0.8 between actual and predictive CI cycles, which is a predictive power of ARIMA(1,0,1) model. Further model persistency estimated by Hurst exponent (smoothness) with the help of fractal dimension (complexity). All predictive CI cycles show persistence and positively correlated, the total dataset equation is given as X(t) – 0.977(t-1) = Z(t) - 0.546(t-1).

太阳活动产生的许多太阳活动，例如黑子，Mg II指数，日冕的日冕指数等，都是由太阳磁场产生的，而太阳磁场既影响着太阳大气，又影响了日光层，包括地球环境。太阳的大气层被光圈识别为可见的太阳圆盘，并位于其上方的各层：色球层，过渡带和电晕。在发射日冕中，太阳活动的性质可以通过太阳活动的日冕指数（CI）来识别。其1944-2009年期间的数据集将用于测试ARIMA模型。基于最小AIC和BIC值，为每个CI周期和总数据集选择ARIMA（1,0,1）模型。所选模型的质量得到（RMSE，MAPE和FPE）的认可。观察到的（原始）CI周期与预测性CI周期的行为显示出很强的相关性，在实际和预测CI周期之间的差异大于0.8，这是ARIMA（1,0,1）模型的预测能力。在分形维数（复杂度）的帮助下，通过Hurst指数（平滑度）估算出进一步的模型持久性。所有预测性CI周期均显示持久性并呈正相关，总数据集方程式为X（t）– 0.977（t-1）= Z（t）-0.546（t-1）。