### Abstract

The hat matrix maps the vector of response values in a regression to its predicted counterpart. The trace of this hat matrix is the workhorse for calculating the effective number of parameters in both parametric and nonparametric regression settings. Drawing on the regression literature, the standard kernel density estimate is transformed to mimic a regression estimate thus allowing extraction of a usable hat matrix for calculating the effective number of parameters of the kernel density estimate. Asymptotic expressions for the trace of this hat matrix are derived under standard regularity conditions for mixed, continuous, and discrete densities. Simulations validate the theoretical contributions. Several empirical examples demonstrate the usefulness of the method suggesting that calculating the effective number of parameters of a kernel density estimator maybe useful in interpreting differences across estimators.

Original language | English (US) |
---|---|

Article number | 106843 |

Journal | Computational Statistics and Data Analysis |

Volume | 143 |

DOIs | |

State | Published - Mar 2020 |

Externally published | Yes |

### Fingerprint

### Keywords

- Degrees of freedom
- Hat matrix
- Matrix trace
- Nonparametric density estimation

### ASJC Scopus subject areas

- Statistics and Probability
- Computational Mathematics
- Computational Theory and Mathematics
- Applied Mathematics

### Cite this

**Determining the Number of Effective Parameters in Kernel Density Estimation.** / McCloud, Nadine; Parmeter, Christopher F.

Research output: Contribution to journal › Article

}

TY - JOUR

T1 - Determining the Number of Effective Parameters in Kernel Density Estimation

AU - McCloud, Nadine

AU - Parmeter, Christopher F.

PY - 2020/3

Y1 - 2020/3

N2 - The hat matrix maps the vector of response values in a regression to its predicted counterpart. The trace of this hat matrix is the workhorse for calculating the effective number of parameters in both parametric and nonparametric regression settings. Drawing on the regression literature, the standard kernel density estimate is transformed to mimic a regression estimate thus allowing extraction of a usable hat matrix for calculating the effective number of parameters of the kernel density estimate. Asymptotic expressions for the trace of this hat matrix are derived under standard regularity conditions for mixed, continuous, and discrete densities. Simulations validate the theoretical contributions. Several empirical examples demonstrate the usefulness of the method suggesting that calculating the effective number of parameters of a kernel density estimator maybe useful in interpreting differences across estimators.

AB - The hat matrix maps the vector of response values in a regression to its predicted counterpart. The trace of this hat matrix is the workhorse for calculating the effective number of parameters in both parametric and nonparametric regression settings. Drawing on the regression literature, the standard kernel density estimate is transformed to mimic a regression estimate thus allowing extraction of a usable hat matrix for calculating the effective number of parameters of the kernel density estimate. Asymptotic expressions for the trace of this hat matrix are derived under standard regularity conditions for mixed, continuous, and discrete densities. Simulations validate the theoretical contributions. Several empirical examples demonstrate the usefulness of the method suggesting that calculating the effective number of parameters of a kernel density estimator maybe useful in interpreting differences across estimators.

KW - Degrees of freedom

KW - Hat matrix

KW - Matrix trace

KW - Nonparametric density estimation

UR - http://www.scopus.com/inward/record.url?scp=85072947954&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85072947954&partnerID=8YFLogxK

U2 - 10.1016/j.csda.2019.106843

DO - 10.1016/j.csda.2019.106843

M3 - Article

AN - SCOPUS:85072947954

VL - 143

JO - Computational Statistics and Data Analysis

JF - Computational Statistics and Data Analysis

SN - 0167-9473

M1 - 106843

ER -