A software package for fitting and assessing multi-dimensional point process models using the R sta- tistical computing environment is described. Methods of residual analysis based on random thinning are discussed and implemented. Features of the software are demonstrated using data on wildfire occurrences in Northern Los Angeles County, California.
In this paper a new method is described for estimating the fire interval distribution of a region using spatial-temporal fire history data. In Los Angeles County, California, detailed information on fires has been available through the use of geographic information systems (GIS) technology. The proposed estimator is applied to GIS data covering the years 1878-1996 and it is shown that fuel age appears to have a nonlinear threshold-type relationship with burn area. The estimator is shown to be more stable than previous estimators and to have good finite and large sample properties.
Writers are often viewed as having an inherent style which can serve as a literary fingerprint. By quantifying relevant features related to literary style, one may hope to classify written works and even attribute authorship to newly discovered texts. Beyond its intrinsic interest, the study of literary styles presents the opportunity to introduce and motivate many standard multivariate statistical techniques. Today the statistical analysis of literary styles is made much simpler by the wealth of real data readily available from the Internet. This paper presents an overview and brief history of the analysis of literary styles. In addition we use canonical discriminant analysis and principal component analysis to identify structure in the data and distinguish authorship.
This paper describes a new method for estimating the fire interval distri- bution of a region using historical wildfire boundary data. The new estimator does not assume a specific model and is shown to have good statistical proper- ties. In Los Angeles County, California, detailed information on wildfires has been made available through the use of geographic information systems (GIS) technology. The proposed estimator is applied to GIS data covering the years 1878–1996 and reveals an apparent nonlinear threshold-type relationship with burn area.
Writers are often viewed as having an inherent style which can serve as a literary fingerprint. By quantifying relevant features related to literary style, one may hope to classify written works and even attribute authorship to newly discovered texts. Beyond its intrinsic interest, the study of literary styles presents the opportunity to introduce and motivate many standard multivariate statistical techniques. Today the statisti- cal analysis of literary styles is made much simpler by the wealth of real data readily available from the Internet. This paper presents an overview and brief history of the analysis of literary styles. In addition we use canonical discriminant analyis and prin- cipal component analysis to identify structure in the data and distinguish authorship.
This paper describes a new method for estimating the fire interval distribution of a region using historical wildfire boundary data. The new procedure does not assume a specific parametric model and can adapt to various types of fire interval relationships. The estimator first averages the proportions of certain age classes which burn across time and weights these proportions by the amount of available fuel. Then, rather than fit known theoretical models, we use local linear smoothing methods to ascertain the overall relationship between fuel age and burning. In Los Angeles County, California, detailed information on wildfires has been made available through the use of geographic information systems (GIS) technology. The proposed procedure is applied to GIS data covering the years 1878 - 1996 and reveals an apparent nonlinear threshold - type relationship with burn area. The result is compared with two well - known parametric models and the relationship appears to conform to the well known Olson model for a fire interval distribution.
The Burning Index (BI) is part of the U.S. National Fire Danger Rating System and is widely used as a tool for fire management and hazard assessment. While the usage of such indices is widespread, assessment of these indices in their repective regions of application is rare. We evaluate the effectiveness of the BI for predicting wildfire occurrences in Los Angeles County, California using space-time point process models. The models are based on an additive decomposition of the conditional intensity, with separate terms to describe spatial and seasonal variability as well as contributions from the BI. The models are fit to wildfire and BI data from the years 1976-2000 using a combination of nonparametric kernel smoothing methods and parametric maximum likelihood. In addition to using AIC to compare competing models, new multi-dimensional residual methods based on approximate random thinning are employed to detect departures from the models and to ascertain the precise contribution of the BI to predicting wildfire occurrence. We find that while the BI appears to have a positive impact on wildfire prediction, the contribution is relatively small after taking into account natural seasonal and spatial variation. In particular, the BI does not appear to take into account increased activity during the years 1979-1981 and can overpredict during the early months of the year.
Cookie SettingseScholarship uses cookies to ensure you have the best experience on our website. You can manage which cookies you want us to use.Our Privacy Statement includes more details on the cookies we use and how we protect your privacy.
