

windowf(timeslab)                            R Documentation

_C_a_l_c_u_l_a_t_e _N_o_n_p_a_r_a_m_t_e_r_i_c _S_p_e_c_t_r_a_l _D_e_n_s_i_t_y _E_s_t_i_m_a_t_e

_D_e_s_c_r_i_p_t_i_o_n_:

     Calculate Nonparamteric Spectral Density Estimate

_U_s_a_g_e_:

     windowf(rho,R0,Q,ioptw,M,n,alpha=0.05)

_A_r_g_u_m_e_n_t_s_:

     rho: Array of length {`M'} (if {`ioptw'} is between 1
          and 5) or length ${t{n}}-1$ if {`ioptw'} is
          between 6 and 8 containing autocorrelations.

      R0: Real scalar containing the sample variance $(>0)$.

       Q: Integer containing the number of frequencies
          between 0 and 1 at which to calculate spectra.

   ioptw: Integer containing the number of the window to be
          used in the estimation procedure as indicated by
          the following:

        : 1 ~~ Truncated periodogram

        : 2 ~~ Bartlett

        : 3 ~~ Tukey

        : 4 ~~ Parzen

        : 5 ~~ Bohman

        : 6 ~~ Daniell

        : 7 ~~ Bartlett-Priestley

        : 8 ~~ Parzen-Cogburn-Davis

       M: Integer $(>0)$ containing scale parameter.

       n: (If either {`ioptw'} is between 6 and 8 or the
          factor for determining confidence intervals is
          desired.) Integer containing the length of the
          data set being analyzed.

   alpha: Real scalar ($0<${`alpha'}$<1$) indicating the
          level of confidence.

_V_a_l_u_e_:

       f: Array of length $[{t{Q}}/2]+1$ containing the
          spectral estimator at the frequencies
          $(j-1)/{t{Q}},j=1,...,[{t{Q}}/2]+1$.

       c: Real scalar variable that can be used to find 95%
          confidence intervals for the true spectral den-
          sity.  The interval at the $i$th frequency would
          be from {`f(i)/c'} to {`f(i)*c'}.

