In this paper, we propose a statistical model to analyze the crosstalk noise induced by process variations on single-walled carbon nanotube (SWCNT). Our approach is based on a closed-form metric, which simplifies the noise peak voltage of SWCNT buses into a linear function of process variations. The methodology is based on an assumption that if the variational parameters are independent Gaussian variables, then the performance of the interconnect also tends to have a normal-distributed character. The experiment results reveal that while being able to save the long computation time of SPICE-based tools, this approach yields little loss in accuracy. Also, it provides a compound result which covers all potentical cases and provides more information, promising it to be much closer to the truth in real manufacture process. Compared with SPICE-based Monte Carlo simulations, the experiments report the error in mean and standard deviation of noise peak to be 1.8% and 4.6%, respectively.