vaso               package:robustbase               R Documentation

_V_a_s_o _C_o_n_s_t_r_i_c_t_i_o_n _D_a_t_a _S_e_t

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

     Finney's data on vaso constriction in the skin of the digits.

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

     data(vaso)

_F_o_r_m_a_t:

     A data frame with 39 observations on the following 3 variables.

     '_V_o_l_u_m_e' Inhaled volume of air

     '_R_a_t_e' Rate of inhalation

     '_Y' vector of 0 or 1 values.

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

     The data taken from Finney (1947) were obtained in a carefully
     controlled study in human physiology where a reflex "vaso
     constriction" may occur in the skin of the digits after taking a
     single deep breath.  The response y is the occurence (y = 1) or
     non-occurence (y = 0) of vaso constriction in the skin of the
     digits of a subject after he or she inhaled a certain volume of
     air at a certain rate.  The responses of three subjects are
     available.  The first contributed 9 responses, the second
     contributed 8 responses, and the third contributed 22 responses.

     Although the data represent repeated measurements, an analysis
     that assumes independent observations may be applied, as claimed
     by Pregibon (1981).

_S_o_u_r_c_e:

     Finney, D.J. (1947) The estimation from individual records of the
     relationship between dose and quantal response. _Biometrika_ *34*,
     320-334

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

     Atkinson, A.C. and Riani, M. (2000) _Robust Diagnostic Regression
     Analysis_, First Edition. New York: Springer, Table A.23.

     Fahrmeir, L. and Tutz, G. (2001) _Multivariate Statistical
     Modelling Based on Generalized Linear Models_, Springer, Table
     4.2.

     Kuensch, H.R., Stefanski, A. and Carrol, R.J. (1989) Conditionally
     unbiased bounded influence estimation in general regression
     models, with applications to generalized linear models, _JASA_
     *84*, 460-466.

     Pregibon, D. (1981) Logistic regression diagnostics, _Annals of
     Statistics_ *9*, 705-724.

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

     data(vaso)
     str(vaso)
     pairs(vaso)

     glmV <- glm(Y ~ log(Volume) + log(Rate), family=binomial, data=vaso)
     summary(glmV)
     ## -->  example(glmrob)  showing classical & robust GLM

