Package 'oddsratio'

Insert odds ratios of GAM(M)s into smoothing function

Description

This function inserts calculated odds ratios of GAM(M)s into a plot of a GAM(M) smoothing function.

Usage

insert_or(
  plot_object = NULL,
  or_object = NULL,
  line_col = "red",
  line_size = 1.2,
  line_type = "solid",
  line_alpha = 1,
  text_alpha = 1,
  text_size = 4,
  text_col = "black",
  rect_alpha = 0.5,
  rect_col = NULL,
  rect = FALSE,
  arrow = TRUE,
  values = TRUE,
  values_yloc = 0,
  values_xloc = NULL,
  or_yloc = 0,
  arrow_length = NULL,
  arrow_yloc = NULL,
  arrow_col = NULL,
  arrow_xloc_r = NULL,
  arrow_xloc_l = NULL
)

Arguments

plot_object	A ggplot object from plot_gam().
or_object	A data.frame as returned from or_gam().
line_col, line_alpha, line_type, line_size	Aesthetics of vertical lines.
text_col, text_alpha, text_size	Aesthetics of inserted values.
rect_col, rect_alpha	Aesthetics of shaded rectangle.
rect	Whether to print a shaded rectangle between the vertical lines.
arrow	Whether to print arrows above the inserted values.
values	Whether to print predictor value information nearby the inserted vertical lines.
values_xloc	x-axis location/shift of values relative to their vertical line. Default to 2\% of x-axis range.
or_yloc, values_yloc	Specifies y-location of inserted odds ratio values. Relative to plotted y-axis range. A positive (negative) value will place the the text higher (lower).
arrow_xloc_r, arrow_xloc_l, arrow_yloc, arrow_length, arrow_col	Axis placement options of inserted arrows. Relative to respective axis ranges.

Details

The idea behind this function is to add calculated odds ratios of fitted GAM models (or_gam()) into a plot showing the smooth function (plot_gam) of the chosen predictor for which the odds ratio was calculated for. Multiple insertions can be made by iterative calling the function (see examples).

Right now the function only accepts inputs from or_gam() objects with slice = FALSE. If you want to insert multiple odds ratio values, call the function multiple times.

Value

ggplot2

See Also

plot_gam(), or_gam()

Examples

library(oddsratio)
library(mgcv)
fit_gam <- gam(y ~ s(x0) + s(I(x1^2)) + s(x2) +
  offset(x3) + x4, data = data_gam) # fit model

# create input objects (plot + odds ratios)
plot_object <- plot_gam(fit_gam, pred = "x2", title = "Predictor 'x2'")
or_object1 <- or_gam(
  data = data_gam, model = fit_gam,
  pred = "x2", values = c(0.099, 0.198)
)

# insert first odds ratios to plot
plot_object <- insert_or(plot_object, or_object1,
  or_yloc = 3,
  values_xloc = 0.04, line_size = 0.5,
  line_type = "dotdash", text_size = 6,
  values_yloc = 0.5, arrow_col = "red"
)

# calculate second odds ratio
or_object2 <- or_gam(
  data = data_gam, model = fit_gam, pred = "x2",
  values = c(0.4, 0.6)
)

# add or_object2 into plot
insert_or(plot_object, or_object2,
  or_yloc = 2.1, values_yloc = 2,
  line_col = "green4", text_col = "black",
  rect_col = "green4", rect_alpha = 0.2,
  line_alpha = 1, line_type = "dashed",
  arrow_xloc_r = 0.01, arrow_xloc_l = -0.01,
  arrow_length = 0.01, rect = TRUE
)

---

Calculate Odds Ratios of Generalized Additive (Mixed) Models

Description

This function calculates odds ratio(s) for specific increment steps of GAM(M) models. Odds ratios can also be calculated for continuous (percentage) increment steps across the whole predictor distribution using slice = TRUE.

Usage

or_gam(
  data = NULL,
  model = NULL,
  pred = NULL,
  values = NULL,
  percentage = NULL,
  slice = FALSE,
  ci = NULL
)

Arguments

data	The data used for model fitting.
model	A fitted GAM(M).
pred	Predictor name for which to calculate the odds ratio.
values	Numeric vector of length two. Predictor values to estimate odds ratio from. Function is written to use the first provided value as the "lower" one, i.e. calculating the odds ratio 'from value1 to value2'. Only used if slice = FALSE.
percentage	Percentage number to split the predictor distribution into. A value of 10 would split the predictor distribution by 10\ Only needed if slice = TRUE.
slice	Whether to calculate odds ratios for fixed increment steps over the whole predictor distribution. See percentage for setting the increment values.
ci	Currently fixed to 95\ Currently supported functions: mgcv::gam, mgcv::gamm, gam::gam. For mgcv::gamm, the model input of or_gam needs to be the gam output (e.g. fit_gam$gam).

Value

A data.frame with (up to) eight columns. perc1 and perc2 are only returned if slice = TRUE:

predictor	Predictor name
value1	First value of odds ratio calculation
value2	Second value of odds ratio calculation
perc1	Percentage value of value1
perc2	Percentage value of value2
oddsratio	Calculated odds ratio(s)
ci_low	Lower (2.5%) confident interval of odds ratio
ci_high	Higher (97.5%) confident interval of odds ratio

See Also

or_glm() plot_gam() insert_or()

Examples

library(oddsratio)
library(mgcv)
fit_gam <- gam(y ~ s(x0) + s(I(x1^2)) + s(x2) +
  offset(x3) + x4, data = data_gam) # fit model

# Calculate OR for specific increment step of continuous variable
or_gam(
  data = data_gam, model = fit_gam, pred = "x2",
  values = c(0.099, 0.198)
)

## Calculate OR for change of indicator variable
or_gam(
  data = data_gam, model = fit_gam, pred = "x4",
  values = c("B", "D")
)

## Calculate ORs for percentage increments of predictor distribution
## (here: 20%)
or_gam(
  data = data_gam, model = fit_gam, pred = "x2",
  percentage = 20, slice = TRUE
)

---

Calculate Odds Ratios of Generalized Linear (Mixed) Models

Description

This function calculates odds ratio(s) for specific increment steps of GLMs.

Usage

or_glm(data, model, incr, ci = 0.95)

Arguments

data	The data used for model fitting.
model	A fitted GLM(M).
incr	Increment values of each predictor given in a named list.
ci	Which confidence interval to calculate. Must be between 0 and 1. Default to 0.95

Details

ci_low and ci_high are only calculated for GLM models because MASS::glmmPQL() does not return confident intervals due to its penalizing behavior.

Currently supported functions: stats::glm,MASS::glmmPQL

Value

A data frame with five columns:

predictor	Predictor name(s)
oddsratio	Calculated odds ratio(s)
ci_low	Lower confident interval of odds ratio
ci_high	Higher confident interval of odds ratio
increment	Increment of the predictor(s)

See Also

or_gam()

Examples

## Example with glm()
library(oddsratio)
# load data (source: http://www.ats.ucla.edu/stat/r/dae/logit.htm) and
# fit model
fit_glm <- glm(admit ~ gre + gpa + rank,
  data = data_glm,
  family = "binomial"
) # fit model

# Calculate OR for specific increment step of continuous variable
or_glm(data = data_glm, model = fit_glm, incr = list(gre = 380, gpa = 5))

# Calculate OR and change the confidence interval level
or_glm(
  data = data_glm, model = fit_glm,
  incr = list(gre = 380, gpa = 5), ci = .70
)

## Example with MASS:glmmPQL()
# load data
library(MASS)
data(bacteria)
fit_glmmPQL <- glmmPQL(y ~ trt + week,
  random = ~ 1 | ID,
  family = binomial, data = bacteria,
  verbose = FALSE
)

# Apply function
or_glm(data = bacteria, model = fit_glmmPQL, incr = list(week = 5))

---

Plot GAM(M) Smoothing Function

Description

Plots the smoothing function of GAM(M) predictors via ggplot2

Usage

plot_gam(
  model = NULL,
  pred = NULL,
  col_line = "blue",
  ci_line_col = "black",
  ci_line_type = "dashed",
  ci_fill = "grey",
  ci_alpha = 0.4,
  ci_line_size = 0.8,
  sm_fun_size = 1.1,
  title = NULL,
  xlab = NULL,
  ylab = NULL,
  limits_y = NULL,
  breaks_y = NULL
)

Arguments

model	A fitted model of class gam.
pred	Predictor name.
col_line	Smoothing function line color.
ci_line_col	Confident interval line color.
ci_line_type	Linetype of confidence interval.
ci_fill	Fill color of area between smoothing function and its confidence interval lines.
ci_alpha	Opacity value of confidence interval.
ci_line_size, sm_fun_size	Line sizes.
title	Plot title.
xlab	x-axis title.
ylab	y-axis title.
limits_y	y-axis limits.
breaks_y	y-axis breaks. Values are handed over to a seq call, e.g. seq(-6, 6, 2).

See Also

or_gam() insert_or()

Examples

library(oddsratio)
library(mgcv)
fit_gam <- mgcv::gam(y ~ s(x0) + s(I(x1^2)) + s(x2) + offset(x3) + x4,
  data = data_gam
)

plot_gam(fit_gam, pred = "x2", title = "Predictor 'x2'")

---