Keywords: Poisson distribution, goodness of fit, spatial pattern
West of Tokyo lies a large alluvial plain, dotted by a network of farming villages. Matui (1968) analysed the position of the 911 houses making up one of those villages. The area studied was a rectangle, 3 km by 4 km. A grid was superimposed over a map of the village, dividing its 12 square kilometres into 1200 plots, each 100 metres on a side. The numbers of houses on each of those plots are recorded in a 30 by 40 matrix of data.
The data are stored in text file with 30 rows and with data columns separated by spaces.
Data File (tab-delimited text)
|Matui, Isamu (1968). Statistical study of the distribution of scattered villages in two regions of the Tonami Plain, Toyama Prefecture. In Spatial Patterns. Edited by Berry and Marble. Prentice-Hall, Englewood Cliff, NJ.|
|Larsen, R.J., and Marx, M.L., An Introduction to Mathematical Statistics and Its Applications 2nd Edition. Prentice-Hall, 1986. Case Study 9.4.1.|
The marginal distribution of the number houses looks like a Poisson distribution. There may however be spatial correlation.
Larsen and Marx (1986) use this data to illustrate a chi-square goodness of fit test of the frequency distribution to the Poisson distribution.