When an APC-frame has been produced by apc.frame, this
function draws a set of estimates from an APC-fit in the frame. An
optional drift parameter can be added to the period parameters and
subtracted from the cohort and age parameters.
# S3 method for apc
lines( x, P, C,
scale = c("log","ln","rates","inc","RR"),
frame.par = options()[["apc.frame.par"]],
drift = 0,
c0 = median( C[,1] ),
a0 = median( A[,1] ),
p0 = c0 + a0,
ci = rep( FALSE, 3 ),
lwd = c(3,1,1),
lty = 1,
col = "black",
type = "l",
knots = FALSE,
... )
apc.lines( x, P, C,
scale = c("log","ln","rates","inc","RR"),
frame.par = options()[["apc.frame.par"]],
drift = 0,
c0 = median( C[,1] ),
a0 = median( A[,1] ),
p0 = c0 + a0,
ci = rep( FALSE, 3 ),
lwd = c(3,1,1),
lty = 1,
col = "black",
type = "l",
knots = FALSE,
... )If an apc-object, (see apc.fit), then
the arguments P, C, c0, a0 and p0
are ignored, and the estimates from x plotted.
Can also be a 4-column matrix with columns age, age-specific
rates, lower and upper c.i., in which case period and cohort effects
are taken from the arguments P and C.
Period effects. Rate-ratios. Same form as for the age-effects.
Cohort effects. Rate-ratios. Same form as for the age-effects.
Are effects given on a log-scale? Character variable, one
of "log", "ln", "rates", "inc",
"RR". If "log" or "ln" it is assumed that
effects are log(rates) and log(RRs) otherwise the actual effects are
assumed given in A, P and C. If A is of
class apc, it is assumed to be "rates".
2-element vector with the cohort-period offset and
RR multiplicator. This will typically be the result from the call of
apc.frame. See this for details.
The drift parameter to be added to the period effect. If
scale="log" this is assumed to be on the log-scale, otherwise
it is assumed to be a multiplicative factor per unit of the first
columns of A, P and C
The cohort where the drift is assumed to be 0; the subtracted
drift effect is drift*(C[,1]-c0).
The age where the drift is assumed to be 0.
The period where the drift is assumed to be 0.
Should confidence interval be drawn. Logical or
character. If character, any occurrence of "a" or "A"
produces confidence intervals for the age-effect. Similarly for
period and cohort.
Line widths for estimates, lower and upper confidence limits.
Linetypes for the three effects.
Colours for the three effects.
What type of lines / points should be used.
Should knots from the model be shown?
Further parameters to be transmitted to points
lines, matpoints or matlines used
for plotting the three sets of curves.
APC.lines returns (invisibly) a list of three matrices of the
effects plotted.
The drawing of three effects in an APC-frame is a rather trivial task, and the main purpose of the utility is to provide a function that easily adds the functionality of adding a drift so that several sets of lines can be easily produced in the same frame.