When the time axis is discretized with a minute constant time width and the stimulation by the event and the reaction of the customer are modeled, the calculation amount becomes remarkably large. Also, models that directly handle continuous time such as survival time analysis are based on the assumption that the covariates are constant, and dynamic events can not be taken into consideration. A replacing unit for replacing each event in an impulse-like event sequence with a plurality of rectangular pulses; a dividing unit for dividing the event sequence generated by replacement at the time of at least one change point of the plurality of rectangular pulses; A state vector generating unit for generating a state vector corresponding to the state of the plurality of rectangular pulses corresponding to each of the plurality of periods, an analysis that performs Poisson regression analysis using a plurality of state vectors corresponding to the plurality of periods And an analyzing unit.
展开▼