

   SSmmookkiinngg DDeeaatthhss AAmmoonngg DDooccttoorrss

   AArrgguummeennttss::

        age: The mid-point of the 10 year age-group for the
             doctors.

      smoke: An indicator of whether the doctors smoked (1) or
             not (0).

          n: The number of person-years in the category.

          y: The number of deaths attributed to coronary artery
             disease.

         ns: The number of smoker years in the category
             (`smoke*n').

   SSUUMMMMAARRYY::

        The `breslow' data frame has 10 rows and 5 columns.

        In 1961 Doll and Hill sent out a questionnaire to all
        men on the British Medical Register enquiring about
        their smoking habits. Almost 70% of such men replied.
        Death certificates were obtained for medical practi-
        tioners and causes of death were assigned on the basis
        of these certificates.  The `breslow' data set contains
        the person-years of observations and deaths from coro-
        nary artery disease accumulated during the first ten
        years of the study.

   DDAATTAA DDEESSCCRRIIPPTTIIOONN::

        This data frame contains the following columns:

   SSOOUURRCCEE::

        The data were obtained from

        Breslow, N.E. (1985) Cohort Analysis in Epidemiology.
        In A Celebration of Statistics A.C. Atkinson and S.E.
        Fienberg (editors), 109-143.  Springer-Verlag.

   OOTTHHEERR RREEFFEERREENNCCEESS::

        Davison, A.C. and Hinkley, D.V. (1997) Bootstrap Meth-
        ods and Their Application. Cambridge University Press.

        Doll, R. and Hill, A.B. (1966) Mortality of British
        doctors in relation to smoking: Observations on coro-
        nary thrombosis.  National Cancer Institute Monograph,
        19, 205-268.

