Abstract
A useful tool while analysing spatial point patterns is the pair correlation function (e.g. Fractals, Random Shapes and Point Fields, Wiley, New York, 1994). In practice, this function is often estimated by some nonparametric procedure such as kernel smoothing, where the smoothing parameter (i.e. bandwidth) is often determined arbitrarily. In this article, a data-driven method for the selection of the bandwidth is proposed. The efficacy of the proposed approach is studied through both simulations and an application to a forest data example.
| Original language | English (US) |
|---|---|
| Pages (from-to) | 336-346 |
| Number of pages | 11 |
| Journal | Scandinavian Journal of Statistics |
| Volume | 34 |
| Issue number | 2 |
| DOIs | |
| State | Published - Jun 2007 |
| Externally published | Yes |
Fingerprint
Keywords
- Composite likelihood
- Pair correlation function
- Spatial point process
ASJC Scopus subject areas
- Mathematics(all)
- Statistics and Probability
Cite this
A composite likelihood cross-validation approach in selecting bandwidth for the estimation of the pair correlation function. / Guan, Yongtao.
In: Scandinavian Journal of Statistics, Vol. 34, No. 2, 06.2007, p. 336-346.Research output: Contribution to journal › Article
}
TY - JOUR
T1 - A composite likelihood cross-validation approach in selecting bandwidth for the estimation of the pair correlation function
AU - Guan, Yongtao
PY - 2007/6
Y1 - 2007/6
N2 - A useful tool while analysing spatial point patterns is the pair correlation function (e.g. Fractals, Random Shapes and Point Fields, Wiley, New York, 1994). In practice, this function is often estimated by some nonparametric procedure such as kernel smoothing, where the smoothing parameter (i.e. bandwidth) is often determined arbitrarily. In this article, a data-driven method for the selection of the bandwidth is proposed. The efficacy of the proposed approach is studied through both simulations and an application to a forest data example.
AB - A useful tool while analysing spatial point patterns is the pair correlation function (e.g. Fractals, Random Shapes and Point Fields, Wiley, New York, 1994). In practice, this function is often estimated by some nonparametric procedure such as kernel smoothing, where the smoothing parameter (i.e. bandwidth) is often determined arbitrarily. In this article, a data-driven method for the selection of the bandwidth is proposed. The efficacy of the proposed approach is studied through both simulations and an application to a forest data example.
KW - Composite likelihood
KW - Pair correlation function
KW - Spatial point process
UR - http://www.scopus.com/inward/record.url?scp=34547382412&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=34547382412&partnerID=8YFLogxK
U2 - 10.1111/j.1467-9469.2006.00533.x
DO - 10.1111/j.1467-9469.2006.00533.x
M3 - Article
AN - SCOPUS:34547382412
VL - 34
SP - 336
EP - 346
JO - Scandinavian Journal of Statistics
JF - Scandinavian Journal of Statistics
SN - 0303-6898
IS - 2
ER -