Perpendicular Distance Models for Line Transect Sampling (Q3747592)

From MaRDI portal

!

This is the item page for this Wikibase entity, intended for internal use and editing purposes. Please use the normal view instead:

scientific article; zbMATH DE number 3984428
Language Label Description Also known as
default for all languages
No label defined
    English
    Perpendicular Distance Models for Line Transect Sampling
    scientific article; zbMATH DE number 3984428

      Statements

      Perpendicular Distance Models for Line Transect Sampling (English)
      0 references
      0 references
      1985
      0 references
      radial distance models
      0 references
      Perpendicular distance line transect models
      0 references
      hazard-rate model
      0 references
      exponential power series
      0 references
      exponential quadratic models
      0 references
      Fourier series estimate
      0 references
      variable-term model
      0 references
      Hermite polynomials
      0 references
      Perpendicular distance line transect models are examined to assess whether any single model can provide a general procedure for analyzing line transect data. Of the two-parameter models considered, the hazard- rate model appears promising, whereas the exponential power series and exponential quadratic models do not. Of the nonparametric models, the Fourier series is the best developed, and is favoured by many researchers as a general model. However, for a given data set, the Fourier series estimate may be highly dependent on the number of terms selected, and so the model is not a clear improvement over the hazard-rate model. A similar variable-term model, using Hermite polynomials, is considered, and is shown to be less dependent on the number of terms selected. There has been some debate about whether the derivative of the density function of perpendicular distances evaluated at 0 should be 0, so that the function has a ''shoulder''. The problem is examined in detail, and it is argued that reliable estimation is not possible from line transect data unless a shoulder exists. Many data sets appear to exhibit no shoulder; possible reasons are examined.
      0 references

      Identifiers