oa CSIR Science Scope - Statistical modelling : an appropriate methodology to deal with uncertainty : statistical modelling
|Article Title||Statistical modelling : an appropriate methodology to deal with uncertainty : statistical modelling|
|© Publisher:||Council for Scientific and Industrial Research (CSIR)|
|Journal||CSIR Science Scope|
|Publication Date||Mar 2008|
|Pages||47 - 48|
In many situations 'true' values of phenomena cannot be observed directly, but only values that are modified by uncontrollable random effects. In other situations, uncertainty arises due to the fact that only a sample of measurements can be obtained from which conclusions about a larger population need to be inferred. Uncertainty is also often present in observed data due to measurement variability. Variations between instruments and the skill of users can, for example, result in differences between measurements of the same phenomenon. In other cases, the observed values represent a stochastic process, such as wind speed or wave height, where repeated realisations will result in different values being observed. Statistical modelling takes account of uncertainty in input data and relates this to the uncertainty in projected outcomes. For these and other situations where uncertainty is present, statistical modelling is often an appropriate methodology.
Article metrics loading...