This method computes the Discrete Wavelet Transform, averages the squares of the coefficients of the transform, and then performs a linear regression on the logarithm of the average, versus the log of j, the scale parameter of the transform. The result should be directly proportional to H.



There are several options available for using this method: method.

  1. The length of the data must be entered (power of 2).
  2. j1, j2 are the beginning and ending octaves for the estimation.
  3. N is the order of the wavelet. (2 default)
  4. plotflag allows plotting or not. (default T)
  5. Calls functions from the Statlib Wavelet package. ( wd, accessD ).
  6. Inside function, a bound.effect is used in the estimation to avoid boundary effects on the coefficients.