#### DMCA

## Published

### Citations

4643 | GG: The measurement of observer agreement for categorical data - JR, Koch - 1977 |

2361 |
Conditional logit analysis of qualitative choice behavior
- McFadden
- 1973
(Show Context)
Citation Context ...tor of estimation parameters and xl are the exogenous variables for all Y and at all locations l. This equation holds, if the error terms are independently and identically distributed as log Weibull (=-=McFadden 1973-=-). Normalizing on all probabilities yields a log-odds ratio (Greene 2000): ⎡ plh ⎤ ln⎢ ⎥ = x'l ( β h − β m ) (3.2.8) ⎣ plm ⎦ The dependent variable is expressed as the log of the odds of one alternati... |

2306 | Applied Logistic Regression - Hosmer, Lemeshow - 2000 |

1624 | Spatial interaction and the statistical analysis of lattice systems - Besag - 1974 |

1508 |
Spatial econometrics: methods and models
- Anselin
- 1988
(Show Context)
Citation Context ...independent variables (trend) are present, or that one or more important regressor variables are missing (Long 1998). The most general formulation of a spatial autoregressive model is Equation 4.2.2 (=-=Anselin 1988-=-; LeSage 1999). y = ρW 1y + Xβ + u u = λW 2u + ε (4.2.2) ε ~ N(0, σ 2 I n) In this equation y contains an nx1 vector of cross-sectional dependent variables, X represents and nxk matrix of explanatory ... |

1323 | The jackknife, the bootstrap, and other resampling plans - Efron - 1982 |

419 |
Location and Land Use.
- Alonso
- 1964
(Show Context)
Citation Context ...port costs and multiple markets, which makes the analysis more complex but the basic insights of the importance of location and transport cost in determining farmer incentives and so land use remain (=-=Alonso 1964-=-; Nelson 2002). The von Thünen model provides a simplified representation of the location decisions that have led to the diverse patterns of land use that cover the earth’s 12ssurface. The complexity ... |

283 | Econometric Analysis, 4 th Edition - Greene - 2000 |

242 |
Spatial Processes, Models, and Applications,
- Cliff, Ord
- 1981
(Show Context)
Citation Context ...ional statistical methods, like linear and logistic regression, in spatial land use analysis is that these methods assume the observations to be statistically independent and identically distributed (=-=Cliff and Ord 1981-=-). However, spatial land use data have the tendency to be dependent, a phenomenon known as spatial autocorrelation. Spatial autocorrelation may be defined as the property of random variables to take v... |

183 |
Evaluating presence–absence models in ecology: the need to account for prevalence
- Manel, Williams, et al.
- 2001
(Show Context)
Citation Context ...6 the area under the curve is computed, where x i is the rate of false positives for scenario i, y i the rate of true positives for scenario i, and n the number of points (Pontius and Schneider 2001; =-=Manel et al. 2001-=-). n AUC =∑ = i 1 [ x − x ][ y + y − y / 2] (3.2.6) i + 1 i i i+ 1 i An ideal model outcome hugs the left side and top side of the graph, and the area under the curve is 1. A random classifier should ... |

178 |
Proximate causes and underlying driving forces of tropical deforestation”, BioScience 52
- Geist, Lambin
- 2002
(Show Context)
Citation Context ...ood extraction or road building. Underlying causes are the ‘fundamental forces’ that underpin the proximate causes, including demographic, economic, technological, institutional and cultural factors (=-=Geist and Lambin 2002-=-). In most cases, a wide range of factors is used to represent the underlying causes; examples include soil suitability, population density, rainfall and accessibility. They can also be differentiated... |

175 | Under the hood. Issues in the specification and interpretation of spatial regression models. - Anselin - 2002 |

152 | The causes of land-use and land-cover change: moving beyond the myths. Global environmental change,
- Lambin, Turner, et al.
- 2001
(Show Context)
Citation Context ...voted to the analysis of relations between land use and the socio-economic and biophysical variables that act as the ‘driving forces’ of land use change (Turner II et al. 1993; Turner II et al. 1995; =-=Lambin et al. 2001-=-). Driving forces are generally subdivided into two groups: proximate causes and underlying causes. Proximate causes are the activities and actions that directly affect land use, e.g. wood extraction ... |

150 | Time-Dependent Analysis for
- Kim, Uysal
- 1998
(Show Context)
Citation Context ...e first chosen factor accounts for most of the variance in the data. The second is chosen in the same way but it has to be orthogonal to the first. The last factor explains all the residual variance (=-=Kim 1970-=-). Common factor analysis is a covariance or correlation-oriented method based on the assumption that each variable is influenced by a set of shared or common factors that determine the correlation be... |

116 |
Dynamics of land-use and land-cover change in tropical regions
- Lambin, Geist, et al.
(Show Context)
Citation Context ...g of the process of LUCC. The three generic approaches to study land use change, which are the empirical, the narrative and the modelling approaches, are representative of the different perspectives (=-=Lambin et al. 2003-=-). Integration of the results of these approaches should lead to a better understanding of LUCC. Techniques for the empirical approach have been extensively discussed in this report. Further implement... |

114 | Roads, land use, and deforestation: a spatial model applied to Belize. World Bank Econ. - Chomitz, Gray - 1996 |

108 | Applied logistic regression analysis (Sage University - Menard - 1995 |

106 |
Multivariate analysis in ecology and systematics: panacea or Pandora’s box?
- James, McCulloch
- 1990
(Show Context)
Citation Context ...elations that have been discovered among the original variables and that these factors are real factors in nature. Unfortunately, factor analysis encourages subjective overinterpretation of the data (=-=James and McCulloch 1990-=-). Example Veldkamp and Fresco (1997) used factor analysis in a LUCC study, in which Costa Rican land use and land cover were investigated at six different scales. Spatial distributions of potential b... |

100 | An Autologistic Model for Spatial Distribution of Wildlife - Augustin, Mugglestone, et al. - 1996 |

93 |
Statistics and data analysis in geology. 2nd ed.
- DAVIS
- 1984
(Show Context)
Citation Context ... In effect, the set of X variables is converted into a single new variable and the set of Y variables into another single new variable. Then the correlation between these new variables is determined (=-=Davis 1986-=-). The correlation coefficients can be interpreted as the square root of the Eigenvalues. Because the correlations pertain to the canonical variates, they are called canonical correlations. Like the E... |

92 |
Numerical Ecology: Developments in Environmental Modeling 20, 2nd Edition.
- Legendre, Legendre
- 1998
(Show Context)
Citation Context ...efined as the property of random variables to take values over distance that are more similar or less similar than expected for randomly associated pairs of observations, due to geographic proximity (=-=Legendre and Legendre 1998-=-). Spatial dependency could be seen as a methodological disadvantage because conventional statistics may lead to the wrong conclusions. On the other hand, the spatial relations actually provide inform... |

91 | Theory, data, methods: developing spatially explicit economic models of land use change.” Agriculture, Ecosystems and Environment - Irwin, Geoghegan - 2001 |

84 |
Multi-agent simulations and ecosystem management: a review.
- Bousquet, Page
- 2004
(Show Context)
Citation Context ...ed models of LUCC are particularly well suited to representing complex spatial interactions under heterogeneous conditions and modelling decentralized, autonomous decision making (Parker et al. 2003; =-=Bousquet and LePage 2004-=-). Multi-agent systems are able to formalize decision-forming behaviour of individual stakeholders, based on a theoretical argumentation. Most multi-agent models focus on either hypothetical or simpli... |

80 |
Comparing discriminant analysis, neural networks, and logistic regression for predicting species distributions: a case study with a Himalayan river bird.
- Manel, Dias, et al.
- 1999
(Show Context)
Citation Context ...the other part of the data. One can use geographic samples, or random samples of the complete data set. This procedure is recommended for testing any presence-absence model on fully independent data (=-=Manel et al. 1999-=-). In ecological studies frequent use has been made of the ROC in applications aiming at prediction. For example, Manel et al. (2001) used this method to predict the occurrence of aquatic invertebrate... |

72 |
The spatiotemporal form of urban growth: measurement, analysis and modeling.
- Herold, Goldstein, et al.
- 2003
(Show Context)
Citation Context ...o cells of the same type). This last index is an overall measure of the landscape heterogeneity and provides a subtle characterization of the spatial arrangement of vacant/undeveloped and urban land (=-=Herold et al. 2003-=-). Another technique for validation of maps is the fuzzy set statistic. The main purpose of the fuzzy set approach is to take into account that there are grades of similarity between pairs of cells in... |

67 |
Statistical Methods for the Social Sciences, (3rd edition)
- Agresti, Finlay
- 1997
(Show Context)
Citation Context ...ion and forward inclusion methods will produce the same results, but when the results differ, backward elimination may uncover relationships missed by forward inclusion, due to the suppressor effect (=-=Agresti and Finlay 1997-=-). This means that in some cases a variable may appear to have a statistically significant effect only when another variable is controlled or held constant. With backward elimination, because both var... |

63 |
Evaluating resource selection functions. Ecological Modelling 157:281–300.
- BOYCE, VERNIER, et al.
- 2002
(Show Context)
Citation Context ...the observed (0 or 1) values of the dependent variable (Menard 2001). Therefore, different methods and measures to evaluate the performance of logistic regression models are used and discussed below (=-=Boyce et al. 2002-=-). Classification table Most statistical programs give a classification table as one of the outputs of a logistic regression. This classification table gives a comparison of observed and predicted val... |

60 |
Do spatial effects really matter in regression analysis
- Anselin, Griffith
- 1988
(Show Context)
Citation Context ...e to be used. The effects of spatial dependence on conventional statistical methods are various, for example biased estimation of error variance, t-test significance levels and overestimation of R 2 (=-=Anselin and Griffith 1988-=-). All the usual statistical tests have the same behaviour: in the presence of positive autocorrelation, computed test statistics are too often declared significant under the null hypothesis. Negative... |

60 | Analysis of land use change: Theoretical and modeling approaches. Lesvos: Regional Research Institute, - Briassoulis - 2000 |

60 |
Applied Multivariate Data Analysis
- Jobson
- 1992
(Show Context)
Citation Context ...the probabilities of local agricultural land use (Gobin et al. 2002) 3.2.3 Multinomial regression Multinomial logit models are used for the case of a dependent variable with more than two categories (=-=Jobson 1992-=-). This type of regression is similar to logistic regression, but it is moresgeneral because the dependent variable is not restricted to two categories. Each category is compared to a reference catego... |

55 | Fuzzy set approach to assessing similarity of categorical maps - Hagen - 2003 |

52 |
The Theory and Practice of Spatial Econometrics,
- LeSage
- 1999
(Show Context)
Citation Context ...riables (trend) are present, or that one or more important regressor variables are missing (Long 1998). The most general formulation of a spatial autoregressive model is Equation 4.2.2 (Anselin 1988; =-=LeSage 1999-=-). y = ρW 1y + Xβ + u u = λW 2u + ε (4.2.2) ε ~ N(0, σ 2 I n) In this equation y contains an nx1 vector of cross-sectional dependent variables, X represents and nxk matrix of explanatory variables, an... |

44 | Foundations of Social Theory. The Belknap - Coleman - 1990 |

43 |
Using neural networks and GIS to forecast land use changes: A land transformational model. Computers, Environments and Urban Systems,
- Pijanowski, Brown, et al.
- 2002
(Show Context)
Citation Context ... networks has increased substantially over the last several years because of the advances in computing performance. A number of applications in land use-related research have been published recently (=-=Pijanowski et al. 2002-=-; Li and Yeh 2002; Pijanowski et al. 2005). Artificial neural networks were developed to model the brain’s interconnected system of neurons so that computers could be made to imitate the brain’s abili... |

40 |
The Uses of Tobit Analysis,’’ Review of Economics and Statistics,
- McDonald, Moffitt
- 1990
(Show Context)
Citation Context ...ation of Tobit analysis is preferred in such cases because it uses both data at the limit (nonadopters – zero values) as well as those above the limit (adopters – positive values) in the estimations (=-=McDonald and Moffit 1980-=-). A direct application of Tobit estimation sufficiently provides the needed information on adoption probability and intensity of use of technologies. It can be shown that the total change in elastici... |

35 | Model goodness of fit: a multiple resolution procedure. - Costanza - 1989 |

35 |
Introduction: the heterogeneity of heterogeneity: a glossary. In:
- Kolasa, Rollo
- 1991
(Show Context)
Citation Context ...s obscure variability whereas fine grain sizes obscure general trends. Shifts in grain size may produce more than averages or constants; they may make homogeneity out of heterogeneity and vice versa (=-=Kolasa and Rollo 1991-=-).s• Emergent properties: changes in grain size are frequently associated with new or emergent properties. In complex, constitutive hierarchies, characteristics of larger units are not simple combinat... |

35 |
Neural-network-based cellular automata for simulating multiple land use changes using GIS
- Li, Yeh
- 2002
(Show Context)
Citation Context ...substantially over the last several years because of the advances in computing performance. A number of applications in land use-related research have been published recently (Pijanowski et al. 2002; =-=Li and Yeh 2002-=-; Pijanowski et al. 2005). Artificial neural networks were developed to model the brain’s interconnected system of neurons so that computers could be made to imitate the brain’s ability to sort patter... |

33 |
Modeling tropical deforestation in the southern Yucatan peninsular region: comparing survey and satellite data.
- Geoghegan, Klepeis, et al.
- 2001
(Show Context)
Citation Context ... linking household and GIS data is to correctly define the spatial observation unit with respect to decision making. In other words: do we link human activities to land or do we link land to people? (=-=Geoghegan et al. 2001-=-). Administrative units or grid cells are not individual agents, but aggregates of them. Inferences as to outcomes in such units require simplifying assumptions about homogeneity of the decision maker... |

30 |
On the evolution of hierarchical urban systems. European Economic Review 43(2):209–251
- Fujita, Krugman, et al.
- 1999
(Show Context)
Citation Context ...; Walker 2004; Walker and Solecki 2004). Other disciplines have also contributed to a theoretical analysis of the processes leading to land use patterns, such as the ‘new’ economic theory of Krugman (=-=Fujita et al. 1999-=-; Krugman 1999). Although these theories do contribute to the explanation of LUCC, there is no single allembracing theory to explain the variety in land use patterns. Therefore empirical methods are f... |

29 | Factors Influencing Adoption of Land-Enhancing Technology in the Sahel: Lessons from a - Baidu-Forson - 1999 |

29 |
The dynamics of landcover change in western Honduras: exploring spatial and temporal complexity.
- Munroe, Southworth, et al.
- 2002
(Show Context)
Citation Context ...etely inaccurate) and 1 (completely accurate) and measures the observed agreement between the classification and the reference data and the agreement that might be attained solely by chance matching (=-=Munroe et al. 2002-=-). The Kappa statistic (Pontius 2002) is: P − P o c κ = (4.3.3) Pp − Pc where P o is the observed proportion correct, P c is the expected proportion correct due to chance and P p is the proportion cor... |

28 | TS (2003). Determinants of land use in Amazônia: A finescale spatial analysis - KM, Thomas |

27 |
The use of backpropagating artificial neural networks in land cover classification,
- Kavzoglu, Mather
- 2003
(Show Context)
Citation Context ...use of artificial neural networks in the classification of remotely sensed image data. 52sFigure 3.7. A simple four-layer fully connected free-forward neural network as used for image classification (=-=Kavzoglu and Mather 2003-=-) Examples Pijanowski et al. (2002) used artificial neural networks to determine the location of land use change using landscape-scale variables, given a certain amount of change determined by regiona... |

27 | Spatial autocorrelation in multi-scale land use models. - Overmars, Koning, et al. - 2003 |

26 | Bayesian spatial probit estimation: a primer and an application to HYV rice adoption - Holloway, Shankar, et al. - 2002 |

23 | Dynamic causal patterns of desertification. - Geist, Lambin - 2004 |

22 | J.A. Agricultural land‑use change in Brazilian Amazonia between 1980 and - CARDILLE, FOLEY - 1995 |

19 | Linking GIS and spatial data analysis in practice - Anselin, Dodson, et al. - 1993 |

19 | An Approach to Linking Remotely Sensed Data and Areal Census Data,” - Chen - 2002 |

19 |
Multi-Agent System models for the simulation of land-use and land-cover change: a review
- PARKER, MANSON, et al.
- 2003
(Show Context)
Citation Context ...ors. Multi-agent-based models of LUCC are particularly well suited to representing complex spatial interactions under heterogeneous conditions and modelling decentralized, autonomous decision making (=-=Parker et al. 2003-=-; Bousquet and LePage 2004). Multi-agent systems are able to formalize decision-forming behaviour of individual stakeholders, based on a theoretical argumentation. Most multi-agent models focus on eit... |

18 |
Role of heterogeneity in scaling of ecological systems under analysis.
- Allen, Hoekstra
- 1991
(Show Context)
Citation Context ... that can be detected in the data. The scale of the study is an interaction of grain and extent. If the extent is large, the sampling protocol will be expensive unless the grain is relatively coarse (=-=Allen and Hoekstra 1991-=-). Each analysis of spatial pattern incorporates scale explicitly or implicitly into the process of identifying research objects: the very act of identifying a particular pattern means that scale, ext... |

18 |
Deforestation and land use change: sparse data environments.
- Nelson, Geoghegan
- 2002
(Show Context)
Citation Context ...hange and is often evaluated both in terms of conventional inference on variable coefficients and goodness of fit, and with respect to ability to predict actual landscape change (Pontius et al. 2004; =-=Nelson and Geoghegan 2002-=-). It should be noted that the same spatial methods are generally equally relevant for farming system and technology choices that are not necessarily closely related to land use, such as changes in ch... |

17 | Introduction to the special issue on spatial analysis for agricultural economists. - Nelson - 2002 |

15 | Stochastic simulation of land-cover change using geostatistics and generalized additive models. - Brown, Goovaerts, et al. - 2002 |

15 |
S+ spatial stats: user’s manual for Windows and Unix.
- Kaluzny, Vega, et al.
- 1998
(Show Context)
Citation Context ...ry variables in the matrix X to explain variation in y over the spatial sample of observations. This model is also called the simultaneous model (Anselin 1988) or simultaneous spatial autoregression (=-=Kaluzny et al. 1997-=-). W 1 = 0 results in a regression model with spatial autocorrelation in the disturbances. Anselin (2002) describes this model as a standard regression model with spatially filtered variables. A model... |

15 |
The Role of Geography
- Krugman
- 1999
(Show Context)
Citation Context ...r and Solecki 2004). Other disciplines have also contributed to a theoretical analysis of the processes leading to land use patterns, such as the ‘new’ economic theory of Krugman (Fujita et al. 1999; =-=Krugman 1999-=-). Although these theories do contribute to the explanation of LUCC, there is no single allembracing theory to explain the variety in land use patterns. Therefore empirical methods are frequently used... |

15 | Predicting land-cover and land-use change in the urban fringe: a case - López, Bocco, et al. - 2001 |

15 |
Meeting in the Middle: The Challenge of Meso-Level Integration,’’ An international workshop on the harmonisation of land-use and land-cover classification,
- Mcconnell, Moran
- 2001
(Show Context)
Citation Context ...ns of other key terms related to scale are listed in Table 2.1. Issues related to scale are of fundamental importance for LUCC studies (Veldkamp et al. 2001; Verburg and Chen 2000; Walsh et al. 2001; =-=McConnell and Moran 2001-=-). Choices concerning scale, extent and resolution critically affect the type of patterns that will be observed, because patterns that appear at one level of resolution or extent may be lost at lower ... |

13 | Modelling land use change with generalized linear models–a multi-model analysis of change between 1860 and 2000 in Gallatin Valley, - Aspinall - 2004 |

13 |
A First Course in Factor Analysis. Lawrence Erlbaum.
- Comrey, Lee
- 1992
(Show Context)
Citation Context ...a terminal solution. This can achieve simpler and more meaningful factor patterns, instead of the highly complex extracted factors that are related to many of the variables rather than to just a few (=-=Comrey and Lee 1992-=-). Prior to factor analysis the collected information on the various variables can be processed. The variables in factor analysis should be quantitative at the interval or ratio level. Categorical dat... |

13 |
Use Scanner: An integrated GIS based model for long term projections of land use in urban and rural areas
- Hilferink, Rietveld
- 1999
(Show Context)
Citation Context ... an example of a LUCC method that uses subpixel information on land use in its countrywide and continental applications. Other examples of LUCC studies that use mixed pixels are the land use scanner (=-=Hilferink and Rietveld 1999-=-) and the ATEAM land use model (Rounsevell et al. 2005). Most LUCC analysis is, however, based on the representation of land use by its dominant land use. Both data representations can theoretically a... |

13 |
Applied Multilevel Analysis. TT-publikaties
- Hox
- 1995
(Show Context)
Citation Context ...g the starting values for the coefficients. The coefficient estimates are then improved using the new variance-covariance matrix, etc. This process continues until a suitable convergence is attained (=-=Hox 1995-=-). In this way results for both the fixed and random parts of the model are estimated more efficiently compared to OLS (Osgood and Smith 1995). There are two principal estimation algorithms, full maxi... |

13 |
Spatial autoregression modeling of site-specific wheat yield.
- Long
- 1998
(Show Context)
Citation Context ...e an autoregressive structure, or that non-linear relationships between the dependent and the independent variables (trend) are present, or that one or more important regressor variables are missing (=-=Long 1998-=-). The most general formulation of a spatial autoregressive model is Equation 4.2.2 (Anselin 1988; LeSage 1999). y = ρW 1y + Xβ + u u = λW 2u + ε (4.2.2) ε ~ N(0, σ 2 I n) In this equation y contains ... |

12 |
An overview of techniques for dealing with large numbers of independent variables in epidemiologic studies. Preventive Veterinary Medicine
- Dohoo, Ducrot, et al.
- 1996
(Show Context)
Citation Context ...egression coefficients greater than 2 should be examined to determine whether collinearity is present (Menard 2001). Collinearity is easy to detect, but there are only few acceptable remedies for it (=-=Dohoo et al. 1997-=-). Deleting a variable involved in collinearity runs the risk of omitted variable bias. Methods to prevent multicollinearity include factor analysis, a priori correlation analysis and stepwise regress... |

12 | Use Dynamics in the Central Highlands of Vietnam: A spatial model combining village survey data with satellite image interpretation - Muller, Zeller |

11 |
Multilevel statistical methods,
- Goldstein
- 1995
(Show Context)
Citation Context ...vels are either spatial, e.g. village territories and districts, or thematic, e.g. the field, household and village levels. The existence of such data hierarchies is neither accidental nor ignorable (=-=Goldstein 1995-=-). To analyse these hierarchical data, multilevel statistics are preferred to the more conventional OLS approach, because with nested data regression coefficients may exhibit dependency, which means t... |

11 | Analysis of land use drivers at the watershed and household level: linking two paradigms at the Philippine forest fringe. - Overmars, Verburg - 2005 |

9 |
GeoDa Tutorials
- Anselin
- 2005
(Show Context)
Citation Context ...aximized log likelihood, the Akaike information criterion (AIC) and the Schwartz criterion (SC). The model with the highest log likelihood, or with the lowest AIC or SC, has the best goodness of fit (=-=Anselin 1992-=-). An example of the comparison of land use change models based on AIC is presented by Aspinall (2004). By rescaling the AIC scores for a series of models against the model with the minimum AIC score ... |

9 | Mapping Livestockoriented agricultural production systems for the developing world. - Kruska, Reid, et al. - 2003 |

9 |
Farm-level models of spatial patterns of land use and land cover dynamics in the Ecuadorian Amazon
- Pan, Walsh, et al.
(Show Context)
Citation Context ...illed in by a woman instead of a man, who is normally the head of a household. One might therefore consider administering separately the household head and his (or her) spouse to acquire information (=-=Pan et al. 2004-=-). Although it may be difficult to adequately represent land cover types on a map, farming systems that include livestock are even more difficult to map (Thornton et al. 2003). Livestock cannot direct... |

8 |
Analysing land-cover changes in relation to environmental variables
- Hietel, Waldhardt, et al.
- 2004
(Show Context)
Citation Context ...scale. Another example of a study that has used canonical correlation analysis is an analysis of the relationship between land cover change trajectories and environmental variables in Hesse, Germany (=-=Hietel et al. 2004-=-). Table 3.2. Standardized canonical coefficients for environmental variables at two scales (Walsh et al. 2001) Scale Variable * Axis 1 Axis 2 Axis 3 30 m NDVI 0.220 0.824 0.132 Elevation 0.945 -0.592... |

8 | Comparing inductive and deductive modeling of land use decisions: Principles, a model and an illustration from the Philippines - Overmas, Groot, et al. |

7 | Tropical deforestation in Madagascar: analyses using hierarchical, spatially explicit, Bayesian regression models. - Agarwal, Silander, et al. - 2005 |

7 |
Applying hierarchical linear modeling to extended longitudinal evaluations: The Boys Town Follow-Up Study
- Osgood, Smith
- 1995
(Show Context)
Citation Context ..., etc. This process continues until a suitable convergence is attained (Hox 1995). In this way results for both the fixed and random parts of the model are estimated more efficiently compared to OLS (=-=Osgood and Smith 1995-=-). There are two principal estimation algorithms, full maximum likelihood and restricted maximum likelihood (Snijders and Bosker 1999). In general restricted maximum likelihood provides more realistic... |

7 | The use of a multilevel statistical model to analyze factors influencing land use: a study of the Ecuadorian Amazon. - Pan, Bilsborrow - 2005 |

6 |
Logistic modeling to derive agricultural land use determinants: A case study from southeastern Nigeria
- Gobin, Campling, et al.
- 2002
(Show Context)
Citation Context ... fallow cultivation Agricultural fields 40 Grass and bush fallow cultivation Bush fallow cultivation Figure 3.5. Partially nested strategy to predict the probabilities of local agricultural land use (=-=Gobin et al. 2002-=-) 3.2.3 Multinomial regression Multinomial logit models are used for the case of a dependent variable with more than two categories (Jobson 1992). This type of regression is similar to logistic regres... |

6 | Package for Social Sciences - Statistical - 2006 |

5 | Linking pastoralists to a heterogeneous landscape: The case of four Maasai group ranches in Kajiado District - BurnSilver, Boone, et al. - 2003 |

5 | Linking Socioeconomic and Remote Sensing Data at the Community or at the Household Level: Two Case Studies from Africa - Lambin - 2003 |

5 |
Scale-dependent errors in the estimation of land - cover proportions; Implications for global land-cover datasets. Photogr. Eng. Remote Sens
- Moody, Woodcock
- 1994
(Show Context)
Citation Context ...es in the land use distribution at coarse scales as some class proportions will diminish and other will increase with scale, depending on the spatial and probability distributions of the cover types (=-=Moody and Woodcock 1994-=-). In applications at the national or continental level land use is represented by designating the relative cover of each land use type in each pixel, e.g. a pixel can contain 30% cultivated land, 40%... |

4 | Ahn TK (2000) The concept of scale and the human dimensions of global change: a survey - GC, Ostrom |

4 |
Statistical Analysis of Land-use Change and Driving
- Hoshino
- 1996
(Show Context)
Citation Context ... normally expected, so in this case the application of canonical correlation analysis was very appropriate. 29sLand use distribution Figure 3.2. Framework of analysis for land use distribution (after =-=Hoshino 1996-=-) Canonical correlation analysis versus factor analysis Canonical correlation and factor analysis both create latent variables (variates) based on a linear combination of measured variables, but facto... |

4 | Multilevel modelling on farmland distribution in Japan. - Hoshino - 2001 |

4 |
Attitudes and farmer conservation behaviour
- Lynne, Shonkwiler, et al.
- 1988
(Show Context)
Citation Context ...it analysis Although logit models (binary variable as dependent variable) are widely used, Tobit models should be preferred when the dependent variable is censored so as to avoid loss of information (=-=Lynne et al. 1988-=-; Holloway et al. 2004). A censored variable is a variable for which values in a certain range are all transformed to (or reported as) a single value (see Box 4 for an example). The regression model t... |

3 |
Tipificación de sistemas de producción agrícola. RIMISP,
- Escobar, Berdegué
- 1990
(Show Context)
Citation Context ...vel of correlation between the variables means that a lot of information is redundant, which confirms the view that typification surveys should contain relatively few questions but many observations (=-=Escobar and Berdegué 1990-=-). The 11 variables were used in the principal component analysis, which resulted in seven factors that explained 85% of the total observed variation. Another example is a study by Hary et al. (1996),... |

3 |
A revised Tobit procedure for mitigating bias in the presence of non-zero censoring with an application to milk-market participation in the Ethiopian highlands
- Holloway, Nicholson, et al.
- 2004
(Show Context)
Citation Context ...h logit models (binary variable as dependent variable) are widely used, Tobit models should be preferred when the dependent variable is censored so as to avoid loss of information (Lynne et al. 1988; =-=Holloway et al. 2004-=-). A censored variable is a variable for which values in a certain range are all transformed to (or reported as) a single value (see Box 4 for an example). The regression model that is based on a cens... |

3 | Land use and road improvements: a spatial econometric analysis. Int. Regional Sci. Rev - Nelson, Stone, et al. - 2003 |

2 | Reducing a spatial database to its effective dimensionality for logistic regression analysis of incidence of livestock disease. Preventive Veterinary Medicine 32 - Duchateau, Kruska, et al. - 1997 |

2 | Land degradation in African pastoral systems and the destocking controversy. Ecolocial Modelling 86 - Hary, Schwartz, et al. - 1996 |

2 | Ndoye O and Lambin EF (2000) “Impact of Macroeconomic Change on Deforestation - Mertens, WD |

2 | Predictive modelling of historical and recent land-use patterns - Peppler-lisbach - 2003 |

1 | Farming systems characterisation in the Kenyan Highlands: Use of alternative methodologies. Working document - Baltenweck, Steeg, et al. - 2004 |

1 | Monteiro A.M., Câmara G., Soares-Filho B.S., Coutinho Cerqueira G. and Lopes Pennachin C. 2003. Stochastic cellular automata modeling of urban land use dynamics: Empirical development and estimation. Computers, Environment and Urban Systems 27:481–509 - Almeida, Batty, et al. |

1 | Aligula H.M., Heitkonig I.M.A. and Reid R.S. 2001. Distribution and diversity of wildlife in northern Kenya in relation to livestock and permanent water points. Biological Conservation 100(3):297–306 - Leeuw, Waweru, et al. |

1 | Improving GIS derived measures of farm market access: An application to milk markets in the East African highlands - Wolff, Staal, et al. - 2000 |

1 | Kimenyi M.S. 2001. Determinants of poverty in Kenya: A household level analysis. Working paper - Geda, Jong, et al. |

1 |
The scope of factor analysis. In: Factor analysis as a statistical method. Butterworths
- Lawley, Maxwell
- 1971
(Show Context)
Citation Context ...orrelation between variables. The implied expectation is that the number of common factors will account for all the observed relations and that such factors will be less than the number of variables (=-=Lawley and Maxwell 1971-=-). In common factor analyses the correlation matrix is transformed before undertaking factor analysis (Kim 1970). In most cases, these two methods yield very similar results. However, PCA is often pre... |

1 | Meta-analysis of agricultural change. In: Moran E.F. and Ostrom E. (eds), Seeing the forest and the trees: Human-environment interactions in forest ecosystems - McConnell, Keys |

1 | Piketty M.G., Lacquies A.E. and Venturieri A. 2002. Crossing spatial analyses and livestock economics to understand deforestation processes in the Brazilian Amazon: The case of São Félix do Xingú in South Pará. Agricultural Economics 27(3):269–294 - Mertens, Poccard-Chapuis |