emplik-internal            package:emplik            R Documentation

_I_n_t_e_r_n_a_l _e_m_p_l_i_k _f_u_n_c_t_i_o_n_s

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

     Internal emplik functions

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

     logelr(x,mu,lam)
     logwelr(x,mu,wt,lam)
     gradf(z,wt,lam)
     llog(z, eps)
     llogp(z, eps)
     llogpp(z, eps)
     Wdataclean2(z,d,wt=rep(1,length(z)))
     Wdataclean3(z,d,zc=rep(1, length(z)),wt=rep(1,length(z)))
     Wdataclean5(z,d,zc=rep(1, length(z)),wt=rep(1,length(z)),xmat)
     DnR(x,d,w,y=rep(-Inf,length(x)))
     solve3.QP(D, d, A, b, meq, factorized=FALSE)
     WKM(x,d,zc=rep(1,length(d)),w=rep(1,length(d)))
     WCY(x,d,zc=rep(1,length(d)),wt=rep(1,length(d)),maxit=25,error=1e-09)
     LTRC(x,d,w=rep(1, length(d)),y=rep(-Inf, length(x)))
     el.test.wt3(x,wt,mu,maxit,gradtol,Hessian,svdtol,itertrace)
     iter(x, y, delta, beta)
     redistF(y, d, Fdist)
     gradf2(lam, funt1, evt1, rsk1, funt2, evt2, rsk2, K, n)
     gradf3(lam, funt1, evt1, rsk1, funt2, evt2, rsk2, K, n)

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

     These are not intended to be called by the user. 

     'Wdataclean2' and 'DnR' are used by the functions 'emplikH1.test',
      'emplikH2.test' and 'emplikdisc.test'.  It is also used by 
     'LTRC'.  They basically cleans the data, sort etc.

     'logelr', 'llog', 'llogp' and 'llogpp'  are used by function
     'el.test'. They are from Owen.

     'Wdataclean2', 'WKM' and 'solve3.QP' are used by function
     'el.cen.test'. WKM is the weighted Kaplan-Meier.

     'WCY' is the weighted Chang and Yang self-consistant estimator for
     doubly censored data.

     'Wdataclean2' is used by 'el.cen.EM'.

     'LTRC' is for Left Truncated and Right Censored data.

     'gradf3' is used by the function emplikHs.test2.

     'el.test.wt3' are similar to 'el.test.wt' and 'el.test.wt2',  but
     can take vector mean as constraint. 'llog', 'llogp' and 'llogpp'
     are used by both 'el.test.wt3' and 'el.test.wt2'. In addition
     'logwelr' is used by 'el.test.wt3'.

     'iter' is for perform one iteration of EM in the Buckley-James
     censored regression estimation. 'redistF' is for redistribution of
     probability, according to Fdist. Used in BJtest().

