vcd (version 0.1-3.5)

Ord.plot: Ord plots

Description

Ord plots for diagnosing discrete distributions.

Usage

Ord.plot(obj, legend = TRUE, estimate = TRUE, tol = 0.1, type = NULL,
         ylim = NULL, xlab = "Number of occurrences",
	 ylab = "Frequency ratio", main = "Ord plot", ...)
Ord.estimate(x, type = NULL, tol = 0.1)

Arguments

obj

either a vector of counts, a 1-way table of frequencies of counts or a data frame or matrix with frequencies in the first column and the corresponding counts in the second column.

legend

logical. Should a legend be plotted?.

estimate

logical. Should the distribution and its parameters be estimated from the data? See details.

tol

tolerance for estimating the distribution. See details.

type

a character string indicating the distribution, must be one of "poisson", "binomial", "nbinomial" or "log-series" or NULL. In the latter case the distribution is estimated from t

ylim

limits for the y axis.

xlab

a label for the x axis.

ylab

a label for the y axis.

main

a title for the plot.

...

further arguments passed to plot.

x

a vector giving intercept and slope for the (fitted) line in the Ord plot.

Value

• a vector giving the intercept and slope of the weighted OLS line.

Details

The Ord plot plots the number of occurrences against a certain frequency ratio (see Friendly (2000) for details) and should give a straight line if the data comes from a poisson, binomial, negative binomial or log-series distribution. The intercept and slope of this straight line conveys information about the underlying distribution.

Ord.plot fits a usual OLS line (black) and a weighted OLS line (red). From the coefficients of the latter the distribution is estimated by Ord.estimate as described in Table 2.10 in Friendly (2000). To judge whether a coefficient is positive or negative a tolerance given by tol is used. If none of the distributions fits well, no parameters are estimated. Be careful with the conclusions from Ord.estimate as it implements just some simple heuristics!

References

J. K. Ord (1967), Graphical methods for a class of discrete distributions, Journal of the Royal Statistical Society, A 130, 232-238.

Michael Friendly (2000), Visualizing Categorical Data. SAS Institute, Cary, NC.

Examples

## Simulated data examples:
dummy <- rnbinom(1000, size = 1.5, prob = 0.8)
Ord.plot(dummy)

## Real data examples:
data(HorseKicks)
data(Federalist)
data(Butterfly)
data(WomenQueue)
par(mfrow = c(2,2))
Ord.plot(HorseKicks, main = "Death by horse kicks")
Ord.plot(Federalist, main = "Instances of 'may' in Federalist papers")
Ord.plot(Butterfly, main = "Butterfly species collected in Malaya")
Ord.plot(WomenQueue, main = "Women in queues of length 10")
par(mfrow = c(1,1))