lzo.test               package:RTisean               R Documentation

_M_o_d_e_l_i_n_g _d_a_t_a _t_r_o_u_g_h _a _z_e_r_o_t_h _o_r_d_e_r _a_n_s_a_t_z

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

     Makes a zeroth order ansatz and estimates the one step prediction
     errors of the model on a multivariate time series.

_U_s_a_g_e:

     lzo.test(series, l, x = 0, m=c(1,2), c, d = 1, n, S = 1, k = 30, r, f = 1.2, s = 1, C)

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

  series: a matrix or a vector.

       l: number of points to use. 

       x: number of lines to be ignored. 

       m: a vector containing the number of components of the time
          series and the embedding dimension.  

       c: a vector containing the columns to be read.

       d: delay for the embedding. 

       n: number of points for which the error should be calculated. 

       S: temporal distance between the reference points. 

       k: minimal numbers of neighbors for the fit. 

       r: neighborhood size to start with. 

       f: factor to increase the neighborhood size if not enough
          neighbors were found. 

       s: steps to be forecasted. 

       C: width of causality window. 

_D_e_t_a_i_l_s:

     The function searches for all neighbors of the point to be
     forecasted and takes as its image the average of the images of the
     neighbors. The given forecast errors are  normalized to the
     standard deviations of each component. In addition to using a
     multicomponent time series, a temporal embedding is possible. 
     That's why the 'm' argument needs two  numbers as input, where the
     first one is the number of components and the second one the
     temporal embedding.

_V_a_l_u_e:

     A matrix of 's' lines, containing the steps forecasted in the
     first column and the normalized forecast errors in the following
     columns for each component of the vector.

_S_e_e _A_l_s_o:

     'predict', 'xzero'.

_E_x_a_m_p_l_e_s:

     ## Not run: 

     dat <- henon(1000)
     zerotherr <- lzo.test(dat, s = 20)
     plot(zerotherr, t="l", xlab= "Steps", ylab= "Normalized error", main = "Zeroth order ansatz prediction errors")

     ## End(Not run)

