nordpred.estimate.object: Nordpred.estimate-object with fit of power5 and poisson...

Description Components Note for S-plus Author(s) References See Also

Description

nordpred uses the power5 and poisson Age-Period-Cohort (APC) models to calculate prediction of cancer incidence and mortality This class of objects is returned by the nordpred.estimate class of functions to represent a fit of power5 and poisson Age-Period-Cohort models for prediction of cancer incidence. Objects of this class have methods print.nordpred.estimate, summary.nordpred and plot.nordpred.

Components

glm

Fitted glm-object

cases

A data.frame with number of cases

pyr

A data.frame with observed and forecasted person years

noperiod

Number of periods used in estimate

gofpvalue

P-value for goodness of fit

startestage

Youngest age group which have been included in the regression model. Predictions for age groups below this limit it based on average rates from the last 10 years.

suggestionrecent

Indicator recommendation build on pvaluerecent for projecting of average trend or use the slope for the last 10 years? If recent = FALSE, recommendation is to use average trend for the whole observation period, and if recent = TRUE recommendation is to use the slope from the last 10 years

pvaluerecent

P-value for use of recent trend based on a significance test for departure from linear trend

linkfunc

Link function used in the model. Default is special version used in the Nordpred project ('power5'), where the link is g(x)=x^0.2, while the alternative is the poisson function ('poisson'), where the link is g(x)=log(x)

The object will also contain the following (see lm): formula, terms, assign and call.

Note for S-plus

Powerlink is made via a special modification in S-PLUS. This works fine for the point estimates, but the variance estimates found via the glm-objects are wrong. For variance estimates, we would rather recommend using R.

Author(s)

Harald Fekjaer and Bjorn Moller (Cancer Registry of Norway)

References

See Also

Other nordpred: is.nordpred.estimate, nordpred.estimate; is.nordpred, nordpred; nordpred.getpred; nordpred.object; nordpred.prediction; plot.nordpred; print.nordpred.estimate; print.nordpred; summary.nordpred


haraldwf/nordpred documentation built on May 17, 2019, 2:41 p.m.