2021
DOI: 10.1007/s11004-021-09941-1
|View full text |Cite
|
Sign up to set email alerts
|

Compositional Scalar-on-Function Regression with Application to Sediment Particle Size Distributions

Help me understand this report

Search citation statements

Order By: Relevance

Paper Sections

Select...
2
1
1
1

Citation Types

0
8
0

Year Published

2022
2022
2024
2024

Publication Types

Select...
6
1

Relationship

1
6

Authors

Journals

citations
Cited by 12 publications
(8 citation statements)
references
References 42 publications
0
8
0
Order By: Relevance
“…6 This is commonly achieved using regression models. In this section, we introduce a compositional scalar-on-function regression model, 25 which provides an appropriate means of including a time-use distribution as an explanatory or predictive variable in a regression model through its characterisation as a PDF and using a ZB-spline representation as described above.…”
Section: Methodsmentioning
confidence: 99%
See 3 more Smart Citations
“…6 This is commonly achieved using regression models. In this section, we introduce a compositional scalar-on-function regression model, 25 which provides an appropriate means of including a time-use distribution as an explanatory or predictive variable in a regression model through its characterisation as a PDF and using a ZB-spline representation as described above.…”
Section: Methodsmentioning
confidence: 99%
“…Hence, CFISA can be described in terms of the basic operations in Bayes spaces as a perturbation of f false~ by a weighting PDF g, which represents a shift of f false~ in the compositional sense. Due to the centreing of the sample f 1 , , f n in SFPCA, it results from the formulation of the functional regression model (15) that β 0 = β ~ 0 I β 1 ( t ) f false¯ ( t ) d t , t I, 25 where β ~ 0 is the intercept from the regression model with the centred functional covariate. After this re-computation, the CFISA model can be expressed as where β ^ 0 and β ^ 1 are the estimates of the regression parameters from the compositional scalar-on-function regression model (17).…”
Section: Compositional Functional Isotemporal Substitution Analysis (...mentioning
confidence: 99%
See 2 more Smart Citations
“…Although this approach provides a promising framework for density-valued response regression, extensions which (additionally) allow for densityvalued explanatory variables remain extremely limited. A first linear Bayes Hilbert space regression model for scalar responses and density-valued covariates was proposed by Talská et al (2021) using a constrained spline representation (Machalová et al, 2021) and extended further by Scimone et al (2021), allowing for both density-valued responses and covariates. While this model allows for linear effects of the functional composition on the response, extensions to generalised (functional) additive models where parts of the predictor are finite or infinite compositions still remains an open topic.…”
Section: Introductionmentioning
confidence: 99%