Model Tuning with Canonical Correlation Analysis

Abstract

Knowledge of the relationship between model parameters and forecast quantities is useful because it can aid in setting the values of the former for the purpose of having a desired effect on the latter. Here it is proposed that a well-established multivariate statistical method known as canonical correlation analysis can be formulated to gauge the strength of that relationship. The method is applied to several model parameters in the Coupled Ocean-Atmosphere Mesoscale Prediction System (COAMPS) for the purpose of ''controlling'' three forecast quantities: 1) convective precipitation, 2) stable precipitation, and 3) snow. It is shown that the model parameters employed here can be set to affect the sum, and the difference between convective and stable precipitation, while keeping snow mostly constant; a different combination of model parameters is shown to mostly affect the difference between stable precipitation and snow, with minimal effect on convective precipitation. In short, the proposed method cannot only capture the complex relationship between model parameters and forecast quantities, it can also be utilized to optimally control certain combinations of the latter.

Canonical correlations in elephant grass for energy purposes

“…The analysis of canonic correlation is a multivariate statistical procedure that allows for verifying the linear multidimensional relationships between two sets of variables (Costa et al, 2011). The association between traits is of great importance for plant breeding, because the selection practiced in a certain trait may cause changes in another (Marzban et al, 2014). This technique has been used in elephant grass (Cunha et al, 2011) to analyse the basal tillers density and plant height which are responsible for production of dry matter of evaluated clones.…”

North Atlantic and Indian Ocean links with Iraq Climate

“…CCA is a multivariate statistical technique that identifies coupled patterns in two sets, with associated time series being maximum correlated [27]. Therefore, relationships between variables are highlighted through this method [28]. The vectors interrelationship may exhibit 'coupled in-phase variability' or, if observations of the x field are made before observations of y, then the former may act as a predictor of the y field, leading to statistical forecast [27].…”

Application of Multivariate Sensitivity Analysis Techniques to AGCM-Simulated Tropical Cyclones

“…3) VARIANCE-BASED SENSITIVITY ANALYSIS Sobol's (1993) variance-based sensitivity analysis has recently been used in atmospheric and climate science (Lee et al 2013;Zhao et al 2013;Marzban et al 2014). It is capable of assessing the relative effect of each of a number of model input parameters, along with their interactions, on one or more model output variables.…”