dlmmc: Dynamical linear model regression for atmospheric time-series analysis

Regression analysis of atmospheric time-series observations is a key endeavour for identifying long-term trends and studying underlying drivers of variability in the data. Multiple linear regression (MLR) has been a standard tool for climate and atmospheric time-series data analysis. However, MLR has a number of well-known shortcomings that can, in the worst case, lead to biased scientific inferences being drawn from the data. Dynamical linear modeling (DLM) provides a more flexible regression framework that addresses a number of the key issues faced by MLR (reviewed below), providing an attractive and more robust alternative.