Time Series Regression model assumptions

Time series models assume that the underlying process that generated the data points is stationary. This means that the process does not change over time and that the long term mean, variance, and autocorrelation structure of the process remain constant. Additionally, the model assumes that the noise variance of the observations remain constant over time, and that there are no outliers or major breaks in the data. Finally, the model assumes that all relevant data were available and taken into account when deriving the model parameters.