Applied Multivariate Statistical Analysis.

Johnson, Richard A.; Wichern, Dean W.

doi:10.2307/2531616

Cited by 3,131 publications

(597 citation statements)

References 1 publication

Supporting

Mentioning

509

Contrasting

Unclassified

Order By: Relevance

“…This necessity was driven by evidence of burning activity in daily images (smoke or active fires observed) or through using CART is an acronym for Classification Trees and Regression Trees and is described by Breiman et al (1984). 4 Fisher's Linear Discriminant Analysis method is described in Johnson and Wichern (1988). 5 SPOT VGT S10 data refers to products that have been composited over a ten-day period following a maximum NDVI criteria.…”

Section: Implementation Strategy Of the Burned Area Algorithmsmentioning

confidence: 99%

A Global Inventory of Burned Areas at 1 Km Resolution for the Year 2000 Derived from Spot Vegetation Data

et al. 2004

View full text Add to dashboard Cite

Biomass burning constitutes a major contribution to global emissions of carbon dioxide, carbon monoxide, methane, greenhouse gases and aerosols. Furthermore, biomass burning has an impact on health, transport, the environment and land use. Vegetation fires are certainly not recent phenomena and the impacts are not always negative. However, evidence suggests that fires are becoming more frequent and there is a large increase in the number of fires being set by humans for a variety of reasons. Knowledge of the interactions and feedbacks between biomass burning, climate and carbon cycling is needed to help the prediction of climate change scenarios. To obtain this knowledge, the scientific community requires, in the first instance, information on the spatial and temporal distribution of biomass burning at the global scale. This paper presents an inventory of burned areas at monthly time periods for the year 2000 at a resolution of 1 kilometer (km) and is available to the scientific community at no cost. The burned area products have been derived from a single source of satellite-derived images, the SPOT VEGETATION S1 1 km product, using algorithms developed and calibrated at regional scales by a network of partners. In this paper, estimates of burned area, number of burn scars and average size of the burn scar are described for each month of the year 2000. The information is reported at the country level. This paper makes a significant contribution to understanding the effect of biomass burning on atmospheric chemistry and the storage and cycling of carbon by constraining one of the main parameters used in the calculation of gas emissions

show abstract

Section: Implementation Strategy Of the Burned Area Algorithmsmentioning

confidence: 99%

A Global Inventory of Burned Areas at 1 Km Resolution for the Year 2000 Derived from Spot Vegetation Data

et al. 2004

View full text Add to dashboard Cite

show abstract

“…PCA [33] is a transformation used for representing high dimensional data in fewer dimensions such that maximum information about the data is present in the transformed space. In the case of LIF spectral data x with dimension N ¼ 714, the objective of PCA was to find the transformation F M such that the new M-dimensional (M < N) random vector y M ¼ F M T x contained the maximum information about x.…”

Section: Linear Pcamentioning

confidence: 99%

Nonlinear pattern recognition for laser‐induced fluorescence diagnosis of cancer

Majumder

Ghosh

Kataria³

et al. 2003

Lasers Surg Med

View full text Add to dashboard Cite

show abstract

“…Logistic regression was also used to model the probability of feeling very stressed as a function of adverse working conditions and individual factors. Model selection proceeded by starting with a full model, then dropping out variables on the basis of likelihood ratio tests [Johnson and Wichern, 1988]. A lack of ®t test [Hosmer and Lemeshow, 1989] was performed on the output of each candidate's logistic regression model for AMCL to assess model ®t.…”

Section: Statistical Analysesmentioning

confidence: 99%