Abstract
A method for estimating the local intensity of a one-dimensional point process is described. The estimator uses an adaptation of Rosenblatt's kernel method of non-parametric probability density estimation, with a correction for end-effects. An expression for the mean squared error is derived on the assumption that the underlying process is a stationary Cox process, and this result is used to suggest a practical method for choosing the value of the smoothing constant. The performance of the estimator is illustrated using simulated data. An application to data on the locations of joints along a coal seam is described. The extension to two-dimensional point processes is noted.
| Original language | English |
|---|---|
| Pages (from-to) | 138-147 |
| Number of pages | 10 |
| Journal | Journal of the Royal Statistical Society Series C: Applied Statistics |
| Volume | 34 |
| Issue number | 2 |
| DOIs | |
| Publication status | Published - 1 Jan 1985 |
| Externally published | Yes |