The engineering that goes into each headset is also rather complex – for the Jabra STONE, in particular, we were faced with fitting square components inside the headset without sacrificing the design.
In order not to penalize Period, we scaled the three covariates, and this is equivalent to giving differential weighting in the constraint during the estimation process, i.e. the simple constraint that b1 + b3 = b2 becomes: (A-6) where b1c, b2c and b3c, are PLSR coefficients when covariates are scaled in the extraction of partial least square components.
The first partial least squares component therefore has the largest covariance with the outcome and the second component has the second largest covariance, etc. PLSR is also a data-dimension reduction method, and usually only the first few components are retained as new covariates, which therefore explain most of the variance in the outcome that can be explained by the original covariates.
To obtain illumination invariance, the descriptor is normalized by the square root of the sum of squared components.
Here, HT is shown to provide relevant information through frequency analysis, which can even be used in multi-component signals, to determine pulse-square components.
As an alternative a functional partial least squares logit regression model is proposed, that has as covariates a set of partial least squares components of the design matrix of the multiple logit model associated to the functional one.
For this example, the optimal number of sparse partial least squares components was K = 3 and the optimal regularization parameter was η = 0.83.
Two different classification methods are used successively to derive prediction rules: NSC and Linear Discriminant Analysis performed on Partial Least Squares components (PLS-LDA).
Classical least squares, partial least squares, principal component regression and inverse least squares are the most common multivariate calibration tool due to their powerful calibration and ease of implementation [25 31].
