ABSTRACT
A fundamental aim of any statistical modelling exercise is to describe, and ideally explain, the variation in a set of data. To achieve this aim, most models combine deterministic and stochastic elements. Statistical inference is the process of drawing formal conclusions from data. Particular forms of inference include testing, estimation and prediction. A statistical model specifies the joint probability distribution for a set of data. Statistical prediction involves making a probability statement about an unobserved random variable. The likelihood function associated with a statistical model is the joint probability distribution of the data considered as a function of the parameters, with the data held fixed at their observed values. In likelihood-based inference, parameters are considered to be constants, whose values are unknown and cannot be measured directly. The computations associated with geostatistical estimation and prediction can be burdensome for large data-sets.
