Wavelet thresholding in partially linear models: a computation and simulation (Q4676853)

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 2169391
Language Label Description Also known as
default for all languages
No label defined
    English
    Wavelet thresholding in partially linear models: a computation and simulation
    scientific article; zbMATH DE number 2169391

      Statements

      Wavelet thresholding in partially linear models: a computation and simulation (English)
      0 references
      0 references
      20 May 2005
      0 references
      partially linear models
      0 references
      soft thresholding
      0 references
      wavelet
      0 references
      universal threshold
      0 references
      minimax threshold
      0 references
      0 references
      0 references
      0 references
      The following problem is considered. Let \(y_1, y_2, \ldots, y_n\) be observed at non-stochastic points \(0 \leq t_1 < \ldots < t_n \leq 1\) of a predictor variable \(t.\) The response and predictor values are connected by a partially linear model \(y_i = x_i' \beta + f(t_i) + \sigma z_i\) for \(i=1, \ldots, n\), where \(x_i\) are fixed known \(p\)-dimensional vectors, \(\beta\) is the unknown \(p\)-dimensional parameter vector, \(f\) is an unknown function, \(\sigma\) is a known constant and \(z_1, \ldots,z_n\) are i.i.d. Gaussian random variables with mean zero and variance one. The parameter vector \(\beta\) and the function \(f\) are to be estimated. The author presents a wavelet thresholding method for estimating the corresponding parameters. A discrete wavelet transform is used for this estimation. A simulation study based on this method is carried out. The simulation results show that the wavelet smoothing approach is comparable to traditional smoothing methods when their assumptions are satisfied. But wavelet smoothing is often superior when the assumptions about the smoothness of the function of the nonparametric part are not satisfied.
      0 references

      Identifiers

      0 references
      0 references
      0 references
      0 references
      0 references
      0 references
      0 references