The test of destination independence proposed by Ginsberg is reexamined. The original derivation of the test requires untestable assumptions about latent failure times. It is shown that these assumptions are unnecessary, although some reservations are expressed as to the utility of the test in social science applications. An alternative formulation is proposed which provides more readily interpretable results. The use of this revised test is illustrated by an application to second home tenure sojourn times.
Get full access to this article
View all access options for this article.
References
1.
BurnettP, 1977, “Tests of a linear learning model of destination choice”Geografiska Annaler59B95–108.
2.
BurnettP, 1978, “Markovian models of movement within urban spatial structures”Geographical Analysis10142–153.
3.
CoxD RLewisP A W, 1966The Statistical Analysis of Series Events (Methuen, Andover, Hants).
4.
DavidH AMoeschbergerM L, 1978The Theory of Competing Risks (Charles Griffin, High Wycombe, Bucks).
5.
DaviesR BO'FarrellP N, 1979Second Home Distributions and Their Evolution over Time in Cemaes, West Wales final report to the Social Science Research Council, Funded Research Project HR5515, obtainable from the British Library, Lending Division, Boston Spa, West Yorks, or SSRC Library, Temple Avenue, London EC4.
6.
GailM, 1975, “A review and critique of some models used in competing risk analysis”Biometrics31209–222.
7.
GinsbergR B, 1978, “The relationship between timing of moves and choice of destination in stochastic models of migration”Environment and Planning A10667–679.
8.
GrossA JClarkV A, 1975Survival Distributions: Reliability Applications in the Biomedical Sciences (John Wiley, New York).
9.
HowardR A, 1971Dynamic Probabilistic Systems, Volume 2: Semi-Markov and Decision Processes (John Wiley, New York).
10.
KalbfleischJ DPrenticeR L, 1980The Statistical Analysis of Failure Time Data (John Wiley, New York).
11.
KendallM GStuartA, 1973The Advanced Theory of Statistics, volume 2 (Charles Griffin, High Wycombe, Bucks).
12.
LeeLThompsonW A, 1974, “Results on failure time and pattern for the series system” in Reliability and Biometry Eds ProschanFSelfringJ (Society for Industrial and Applied Mathematics, Philadelphia, PA) pp 291–302.
PrenticeR LKalbfleischJ DPetersonA VFlournoyNFarewellV TBreslowN E, 1978, “The analysis of failure times in the presence of competing risks”Biometrics34541–554.
17.
SingerBSpilermanS, 1979, “Clustering on the main diagonal in mobility matrices” in Sociological Methodology 1979 Ed. SchuesslerK F (Jossey-Bass, San Francisco, CA) pp 172–208.
18.
TsiatisA, 1975, “A non-identifiability aspect of the problem of competing risks”Proceedings of the National Academy of Sciences of the United States of America7220–22.
19.
WrigleyN, 1980, “An approach to the modelling of shop choice patterns: An exploratory analysis of purchasing patterns in a British city” in Geography and the Urban Environment, volume III Eds HerbertD TJohnstonR J (John Wiley, Chichester, Sussex) pp 45–85.
