Subspaces and their corresponding reproducing kernels for the RKHS,
Abstract
Smoothing spline models have shown to be effective in various fields (e.g., engineering and biomedical sciences) for understanding complex signals from noisy data. As nonparametric models, smoothing spline ANOVA (Analysis Of variance) models do not fix the structure of the regression function, leading to more flexible model estimates (e.g., linear or nonlinear estimates). The functional ANOVA decomposition of the regression function estimates offers interpretable results that describe the relationship between the outcome variable, and the main and interaction effects of different covariates/predictors. However, smoothing spline ANOVA (SS-ANOVA) models suffer from high computational costs, with a computational complexity of ON3 for N observations. Various numerical approaches can address this problem. In this chapter, we focus on the introduction to a state space representation of SS-ANOVA models. The estimation algorithms based on the Kalman filter are implemented within the SS-ANOVA framework using the state space representation, reducing the computational costs significantly.
Keywords
- Kalman filter
- smoothing spline
- functional ANOVA
- state space representation
- Markov structure
1. Introduction
Smoothing spline ANOVA (SS-ANOVA) has been widely used in various applications [1, 2, 3]. The representer theorem enables an exact solution of regression function in SS-ANOVA models by minimizing a regularized function in a finite-dimensional space, even though the problem resides in a infinite-dimensional space. While SS-ANOVA models have strong theoretical properties, the estimation algorithms used to fit these models are computational intensive, with a computational complexity of
In this book chapter, we focus on the estimation approaches based on the Kalman filter. The Kalman filter was originally created to solve linear filtering and prediction problems used to generate simulations for the Apollo 11 project [8]. More recently, it has been implemented in a variety of engineering and biomedical fields [9, 10]. The Kalman filter is naturally used to fit state space models, methods that use recursive calculations on each observation entered one at a time and resulting in calculations on more accurate unknown variables after each iteration. The Kalman filter updates the state of the dynamic system given a new observation based on the state after the previous observation and the information gained from the new observation. This memory-less property and its simple recursive formulas make Kalman filter approaches computationally efficient, making them a useful tool for big data analytics. SS-ANOVA models can be reformulated to a state space representation, allowing computationally efficient Kalman filter-based model fitting and reducing computational costs to
Section 2 of this chapter will provide the theoretical background of SS-ANOVA models. Section 3 provides a brief background on state space models. The state space representation of SS-ANOVA models can be found in Section 4, along with a simulation study under the univariate setting in Section 5. Section 6 concludes this chapter.
2. Smoothing spline ANOVA models
We assume the data
where
where the first term measures the goodness of fit of
2.1 ANOVA
2.1.1 Classical ANOVA
Classical ANOVA can be used to help to understand the decomposition of regression function in (1). We use a one-way classical ANOVA model as an example. The outcome
where
where
2.1.2 Functional ANOVA
The multivariate function
Theorem 1.1 If
Theorem 1.1 implies that the RKHS
where
In general, the inner product in
where
where
2.2 An example of RKHS on 0 1 2
We will use one example of an RKHS on
where
Now consider the RKHS
where the first four terms in (9) are in the null space
Subspace | Reproducing Kernel |
---|---|
2.3 Estimation
The estimation algorithm of SS-ANOVA models relies on the following representer theorem.
Theorem 1.2 (Representer Theorem) There exist coefficient vectors
where
Taking into consideration model (1), the function
where
We differentiate (12) with respect to
For given smoothing parameters, solving for
2.4 Selection of smoothing parameters
The smoothing parameters
To avoid overparameterization, let
where
2.5 Computational complexity
In this section, we will discuss the computation complexity for calculating
3. State space models
State space methodology was traditionally used to study dynamic problems (e.g., space tracking settings) because the procedure allows for “real-time” updating as data are collected [20]. In this chapter, we use the linear Gaussian state space model as an example to introduce concepts of state space models using the Kalman filter approach. More applications of state space models can be found in a study by Douc, Moulines and Stoffer (2014) and Durbin and Koopman (2001) [21, 22]. A state space model consists of two equations: state equation and measurement equation. The state equation describes the dynamics of the state variables:
where
where
These two vectors are also uncorrelated to the initial state vector
The Kalman filter utilizes a set of recursive equations to estimate
where
and
Then the prediction error vector
with the covariance matrix
Using the facts of the joint distribution of
We further define Kalman gain as
Then the filtered estimate of
4. State space representation of SS-ANOVA models
Due to the Markov structure of SS-ANOVA models after reparameterization, the SS-ANOVA models can be represented by the state space models, allowing for efficient estimation by algorithms based on the Kalman filter. Wecker and Ansley (1983) showed the state space representation for univariate smoothing spline models [11]. Such an approach reduces the computational complexity of smoothing splines from
4.1 Univariate setting
We can use the state space formulation to represent model (1) (
where the covariate
where
The vector
The state space formulation relies on the Markov structure of
We can easily verify
To show the Markov structure of
For any
Thus we have
We now apply (30) and (33) to obtain the Markov structure of
The Markov structure is the key to the state space representation of SS-ANOVA models. For
as the measurement equation, where
for
5. Simulation studies
We used simulated data to compare the estimates of smoothing splines with those based on state space representation under the univariate setting. The following model was used to simulate
where
for
where
6. Conclusions
In this chapter, we have introduced the theoretical foundation (e.g., representer theorem) and estimation algorithms of SS-ANOVA models. Given tensor product operations, the SS-ANOVA models can handle the multivariate data and study the main and interaction effects in the corresponding subspaces via functional ANOVA. The estimation algorithms of SS-ANOVA models need
Acknowledgments
Sun’s research was supported by the 18th Mile TRIF Funding from the University of Arizona and the NIH grant U19AG065169. LaFleur’s research was supported by the NIH grant U19AG065169.
References
- 1.
Brinkman ND. Ethanol fuel-single-cylinder engine study of efficiency and exhaust emissions. SAE Transactions. 1981; 90 :1410-1424 - 2.
Gu C. Smoothing Spline ANOVA Models. New York, NY: Springer; 2013 - 3.
Wahba G. Spline Models for Observational Data. Philadelphia, PA: Society for Industrial and Applied Mathematics; 1990 - 4.
Kim YJ, Gu C. Smoothing spline Gaussian regression: More scalable computation via efficient approximation. Journal of the Royal Statistical Society: Series B (Statistical Methodology). 2004; 66 (2):337-356 - 5.
Ma P, Huang JZ, Zhang N. Efficient computation of smoothing splines via adaptive basis sampling. Biometrika. 2015; 102 (3):631-645 - 6.
Helwig NE, Ma P. Smoothing spline ANOVA for super-large samples: Scalable computation via rounding parameters. Statistics and Its Interface. 2016; 9 :433-444 - 7.
Sun X, Zhong W, Ma P. An asymptotic and empirical smoothing parameters selection method for smoothing spline ANOVA models in large samples. Biometrika. 2021; 108 (1):149-166 - 8.
Kalman RE. A new approach to linear filtering and prediction problems. Journal of Basic Engineering. 1960; 82 (1):35-45 - 9.
Moreno VM, Pigazo A. Kalman Filter: Recent Advances and Applications. London, UK: IntechOpen; 2009 - 10.
Shumway RH. Longitudinal data with serial correlation: A state-space approach. Journal of Statistical Planning and Inference. 1995; 47 (3):395-397 - 11.
Wecker WE, Ansley CF. The signal extraction approach to nonlinear regression and spline smoothing. Journal of the American Statistical Association. 1983; 78 (381):81-89 - 12.
Qin L, Guo W. State space representation for smoothing spline ANOVA models. Journal of Computational and Graphical Statistics. 2006; 15 (4):830-847 - 13.
Koopman SJ, Durbin J. Fast filtering and smoothing for multivariate state space models. Journal of Time Series Analysis. 2000; 21 (3):281-296 - 14.
Wang Y. Smoothing Splines: Methods and Applications. Boca Raton, FL: CRC Press; 2011 - 15.
Yang Y. Model selection for nonparametric regression. Statistica Sinica. 1999; 9 :475-499 - 16.
Abramowitz M, Stegun IA. Handbook of Mathematical Functions with Formulas, Graphs, and Mathematical Tables. Washington, DC: U.S. Govt. Print. Off; 1964 - 17.
Wahba G. A comparison of GCV and GML for choosing the smoothing parameter in the generalized spline smoothing problem. The Annals of Statistics. 1985; 13 :1378-1402 - 18.
Xiang D, Wahba G. A generalized approximate cross validation for smoothing splines with non-Gaussian data. Statistica Sinica. 1996:675-692 - 19.
Graven P, Wahba G. Smoothing noisy data with spline function: Estimating the correct degree of smoothing by the method of Generalized Cross-Validaton. Numerische Mathematik. 1979; 31 :377-403 - 20.
De Jong P. The likelihood for a state space model. Biometrika. 1988; 75 (1):165-169 - 21.
Douc R, Moulines E, Stoffer D. Nonlinear Time Series: Theory, Methods and Applications with R Examples. Boca Raton, FL: CRC Press; 2014 - 22.
Durbin J, Koopman SJ. A Simple and Efficient Simulation Smoother for State Space Time Series Analysis. Biometrika, 2002; 89 (3);603-615 - 23.
Wahba G. Improper priors, spline smoothing and the problem of guarding against model errors in regression. Journal of the Royal Statistical Society: Series B (Methodological). 1978; 40 (3):364-372 - 24.
Gu C. Smoothing Spline ANOVA Models: R Package gss. Journal of Statistical Software. 2014; 58 (5):1-25