JLLp                 package:JLLprod                 R Documentation

_P_a_r_a_m_e_t_r_i_c _G_e_n_e_r_a_l_i_z_e_d _H_o_m_o_t_h_e_t_i_c _P_r_o_d_u_c_t_i_o_n _F_u_n_c_t_i_o_n

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

     This function simply fits 2 parametric models, P2 and P3, as
     described in Jacho-Chvez, Lewbel and Linton (2005).

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

     JLLp(lnY,lnK,lnL,theta,model)

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

     lnY: Ln of Y, output.

     lnK: Ln of K, capital.

     lnL: Ln of L, labour.

   theta: A list of starting values of the form list(a,b0,b1,b2,g) if
          model=2 or a list of the form list(a,b0,b1,b2) if model=3.

   model: Scalar, 2 or 3.

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

     This function is a simple call to `nls' to fit specific parametric
     models for production by NonLinear Least Squares.

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

     A `nls' object.

_W_a_r_n_i_n_g:

     This function is very sensible to theta. It may fail most of the
     time. If this is the case, the user is adviced to use their own
     call to `nls' or `optim' which is most likely to work.

_A_u_t_h_o_r(_s):

     David Toms Jacho-Chvez

_R_e_f_e_r_e_n_c_e_s:

     Jacho-Chvez, D.T., Lewbel, A., and Linton, O.B. (2005)
     Identification and Nonparametric Estimation of a Transformed
     Additively Separable Model. Unpublished manuscript.

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

     'JLL'

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

     library(JLLprod)
     data(ecu)
     ##This part simply does some data sorting & trimming
     xlnK <- ecu$lnk
     xlnL <- ecu$lnl
     xlnY <- ecu$lny
     xqKL <- quantile(xlnK-xlnL,  probs=c(2.5,97.5)/100)
     yx <- cbind(xlnY,xlnK,xlnL)
     tlnklnl <- yx[((yx[,2]-yx[,3])>=xqKL[1]) & ((yx[,2]-yx[,3])<=xqKL[2]),]
     tlnklnl[,2]<-tlnklnl[,2]-tlnklnl[,3]

     bb<-list(b0=11,b1=1,b2=0,g=-0.15,a=0.4)
     Y <- tlnklnl[,1]; K <- tlnklnl[,2]; L <- tlnklnl[,3]
     pJLL<-JLLp(Y,K,L,theta=bb,model=2)
     print(summary(pJLL))

