Published online by Cambridge University Press: 04 January 2017
Recent interest in voters' anti-incumbent sentiments focuses on generational change as well as public weariness with partisan control of a long duration. Theories on the electoral effects of such behaviors predict partisan cycles that suggest rising hazards of party incumbency. This article provides an analytical framework for examining changes and durations of party control in presidential elections as a discrete point process. We introduce the discrete Weibull distribution for testing contagion in the context of renewal theory and develop the notion of pseudo-periodicity for a binary process. Our findings based on this event history approach confirm the claim that party incumbency engenders rising hazards. The partisan cycles that we identified have a pseudo-period of approximately six to eight presidential terms.