season (version 0.3.15)

monthglm: Fit a GLM with Month

Description

Fit a generalized linear model with a categorical variable of month.

Usage

monthglm(
  formula,
  data,
  family = gaussian(),
  refmonth = 1,
  monthvar = "month",
  offsetmonth = FALSE,
  offsetpop = NULL
)

Arguments

formula

regression model formula, e.g., y~x1+x2, (do not add month to the regression equation, it will be added automatically).

data

a data frame.

family

a description of the error distribution and link function to be used in the model (default=gaussian()). (See family for details of family functions.).

refmonth

reference month, must be between 1 and 12 (default=1 for January).

monthvar

name of the month variable which is either an integer (1 to 12) or a character or factor (`Jan' to `Dec' or `January' to `December') (default='month').

offsetmonth

include an offset to account for the uneven number of days in the month (TRUE/FALSE). Should be used for monthly counts (with family=poisson()).

offsetpop

include an offset for the population (optional), this should be a variable in the data frame. Do not log-transform the offset as the log-transform is applied by the function.

Value

call

the original call to the monthglm function.

fit

GLM model.

fitted

fitted values.

residuals

residuals.

out

details on the monthly estimates.

Details

Month is fitted as a categorical variable as part of a generalized linear model. Other independent variables can be added to the right-hand side of formula.

This model is useful for examining non-sinusoidal seasonal patterns. For sinusoidal seasonal patterns see cosinor.

The data frame should contain the integer months and the year as a 4 digit number. These are used to calculate the number of days in each month accounting for leap years.

References

Barnett, A.G., Dobson, A.J. (2010) Analysing Seasonal Health Data. Springer.

See Also

summary.monthglm, plot.monthglm

Examples

# NOT RUN {
data(CVD)
mmodel = monthglm(formula=cvd~1 ,data=CVD, family=poisson(),
                  offsetpop=expression(pop/100000), offsetmonth=TRUE)
summary(mmodel)

# }