With the diversification of paid search advertising channels, e-commerce enterprises are paying more and more attention on how to evaluate the effectiveness of different paid search advertising channels correctly and accurately to choose the optimal advertising channel or channels. We develop a multivariate time series model to investigate the spillover effect of paid search advertising channels based on the ad click-through rate and conversion rate, and calibrate the model using an e-commerce site's web log data. We determine the long-term equilibrium relationship between each channel's advertisement clicks through the co-integration test and evaluate the effect of short-term fluctuations in the interaction between each channel advertisement clicks through the vector error correction model. Based on the empirical results, this paper puts forward suggestions on the advertising strategy of this e-commerce website.