Spatial Data Analysis of Regional Counts

Abstract

Summaryhdemie-Verlag Berlin h u n t s data from spatially conthguoue regions offer a challenge to the statistician both from the data snalytic and the statistical modeling point of view. Important applications include epidemiological studies (e. g., cancer mortality over the counties of the USA) and Census surveys (e. g., undercount over the Census blocks of an urban ares). It has long been recognized by time-series anelysta that data close together in time usually exhibit higher dependence than those far apart. Time-series data analysis relies on methods of data transformation, detrending, and autocorrelation plotting. It is our intention in this article to generalize this approach to a spatial setting. To do this we consider 8 smell spatial deb set of 100 observations. Through the use of 8 square-root transformation, a weighted median polish and a variogram analysis of the median-polish residuals, we represent the transformed data 88 a trend plus stationary error. Thus we show how standard dab-analytic techniques can be modified both to mitigate and to exploit the spatial relationships.

Análise da distribuição espacial da mortalidade neonatal e de fatores associados, em Salvador, Bahia, Brasil, no período 2000-2006

“…Nessa perspectiva, este estudo teve como objetivos identificar padrões na distribuição espacial e áreas de risco para a mortalidade neonatal, bem como, examinar possíveis associações entre fatores relativos à atenção à saúde materno-infantil, características maternas e condições de nascimento, e a variação espacial desta mortalidade, em Salvador, Bahia, no período de 2000-2006. [10][11][12][13][14][15][16][17][18][19] anos, e proporção de nascidos vivos de mães com idade igual ou superior a 35 anos); condições de nascimento (proporção de nascidos vivos com peso inferior a 2.500g e proporção de nascidos vivos com idade gestacional abaixo de 37 semanas); atenção à saúde materno-infantil (proporção de nascidos vivos de mães com número de consultas pré-natais inadequado à idade gestacional, proporção de nascidos vivos de partos cesáreos, e proporção de nascimentos vivos em hospitais públicos). "Proporção de mães com escolaridade inadequada à idade" e "proporção de nascidos vivos de mães com número de consultas pré-natais inadequado à idade gestacional" foram calculadas conforme preconizado pelo Ministério da Educação para escolarização e número de consultas recomendado pelo Ministério da Saúde 16 para cada trimestre da gestação, respectivamente.…”

Bacterial meningitis and living conditions

“…Next, the incidence and fatality rates for all bacterial forms of meningitis were calculated for each IZ for that period to minimize possible distortions resulting from the small number of cases and deaths from the disease in the various IZs. The spatial distribution pattern of meningitis was identified by inspecting thematic maps constructed using ArcView 3.3 and the spatial statistics module of S-PLUS 6.0 based on the distribution of the crude and adjusted rates of incidence according to IZ, following Freeman-Tukey transformation 30 . Next, the existence of spatial autocorrelation of the incidence rate was evaluated using the global Moran index 31 , after excluding 2 IZs that consisted of islands belonging to the municipality.…”

Smoothing Regional Maps Using Empirical Bayes Predictors

“…However, in this analysis of the standardized data {Yi} given by (36), Anson county's value was large but not unusually so. Hence all one hundred counties were included in the analysis to Based on a spatial data analysis of the type described in Cressie and Read (1989) and Cressie and Chan (1989), X was chosen to be an n x 2 matrix whose first column is the vector of Is, and whose second column is the vector of (FreemanTukey transformed) nonwhite live birth rates. The matrix r was chosen as follows: 92 I Geographical Analysis their non-Bayesian spatial analysis); the transformed standardized SID rate given by (36); the empirical Bayes predictor given by (12) with 8 = 8; and the empirical Bayes predictor's estimated root mean squared prediction error given by the square root of the diagonal elements of (33).…”

“….. . Cressie and Read [1989]), the empirical Bayes predictor possesses statistical optimality properties that make it a very attractive smoother. Moreover, each smoothed value is accompanied by a measure of uncertainty, namely, its mean squared prediction error (whose estimation is discussed in section 4).…”

“…These issues are raised by Cressie and Read (1989), who analyze SID rates for the counties of North Carolina. Wallin (1984) discusses inter alia the problem of mapping ratios and identifies a need for an analysis of their uncertainty and variability.…”