A comparison of continuous and discrete time modeling of affective processes in terms of predictive accuracy

Abstract Intra-individual processes are thought to continuously unfold across time. For equally spaced time intervals, the discrete-time lag-1 vector autoregressive (VAR(1)) model and continuous-time Ornstein–Uhlenbeck (OU) equivalent. It is expected that by taking into account unequal spacings of intervals in real data between observations will lead an advantage for OU terms predictive accuracy. In this paper, claim being investigated comparing accuracy VAR(1) on typical ESM obtained context affect research. shown outperforms majority series, even though unequally spaced. Accounting measurement error does not change result. Deleting large abrupt changes short (that may be caused externally driven events) however a significant improvement model. This suggests psychology evolving, but there factors, like external events, which can disrupt continuous flow.