STOCHASTIC DISTURBANCE SPECIFICATION AND ITS IMPLICATIONS FOR LOG-LINEAR REGRESSION

被引：27

作者：

HAWORTH, JM ^{[1
]}

VINCENT, PJ ^{[1
]}

机构：

[1] UNIV LANCASTER,DEPT GEOG,LANCASTER LA1 4YW,ENGLAND

来源：

ENVIRONMENT AND PLANNING A | 1979年 / 11卷 / 07期

关键词：

D O I：

10.1068/a110781

中图分类号：

X [环境科学、安全科学];

学科分类号：

08 ; 0830 ;

摘要：

Model specification is a crucial factor in regression. In log-linear models, although the estimated relationship represents the conditional expectation of ln Y, the antilogarithm does not give the conditional expectation of Y. This has implications in spatial analysis. -Authors

引用

页码：781 / 790

页数：10

共 18 条

[1] Aitchison J, 1957, LOGNORMAL DISTRIBUTI
[2] BAXTER RS, 1976, COMPUTER STATISTICAL
[3] Cliff A.D., 1973, SPATIAL AUTOCORRELAT
[4] DRAPER N, 1966, APPLIED REGRESSION A
[5] EBDON D, 1977, STATISTICS GEOGRAPHY
[6] FERGUSON RI, 1977, CONCEPTS TECHNIQUES, V15
[7] Goldberger AS, 1968, ECONOMETRICA, V35, P464
[8] Goldfeld SM, 1972, NONLINEAR METHODS EC
[9] ACQUIRING SPATIAL INFORMATION
GOULD, P
[J]. ECONOMIC GEOGRAPHY, 1975, 51 (02) : 87 - 99
[10] IS STATISTIX-INFERENS GEOGRAPHICAL NAME FOR A WILD GOOSE
GOULD, P
[J]. ECONOMIC GEOGRAPHY, 1970, 46 (02) : 439 - 448

← 1 2 →