Skip to contents

Concepts and objects

Start here. Background on kernels, fitting methods, and predictions, a glossary of terms, and descriptions of the objects the main functions return.

multiScaleR multiScaleR-package
multiScaleR: estimate scales of effect for spatial covariates
multiScaleR-glossary
Glossary of terms used in multiScaleR
multiScaleR-kernels
Kernels and scales of effect
multiScaleR-fitting-methods
Fitting methods for estimating scales of effect
multiScaleR-prediction
Prediction quantities and their uncertainty
multiScaleR-class
Fitted multiScaleR objects
multiScaleR_data-class
Prepared inputs for scale estimation
multiScaleR_bayes-class
Bayesian landscape-scale fits
multiScaleR_cv-class multiScaleR_choice_cv-class multiScaleR_field_cv-class multiScaleR_mixed_cv-class multiScaleR_bayes_cv-class
Cross-validation results

Prepare inputs

Build the kernel inputs and declare how each raster covariate is summarized before optimization.

kernel_prep()
Prepare Kernel Inputs for Scale Optimization
kernel_prep_by_group()
Prepare Kernel Inputs from Group-Specific Rasters
msr_vars() kernel_var() landscape_var() surface_var() print(<multiScaleR_vars>)
Define Model Covariates from Raster Layers
prepare_unmarked_data()
Prepare spatial covariates for native unmarked prediction
validate_unmarked_data() print(<msr_unmarked_audit>)
Audit Survey Identities and Paired Distance-Removal Counts
estimate_multiscale_ram() print(<multiScaleR_ram>)
Estimate Memory Requirements for Multiscale Optimization

Optimize scales of effect

Estimate the scale of effect for each covariate by maximum likelihood and assess the fit.

multiScale_optim()
Estimate scales of effect by maximum likelihood
profile_sigma()
Profile Model Fit Across Sigma
scale_inference()
Profile and Bootstrap Inference for an NB2 Scale
diagnostics()
Retrieve diagnostics from multiScaleR objects
vcov(<multiScaleR>)
Covariance matrices for an analytic multiscale fit

Bayesian scale estimation

Estimate scales of effect with full posterior uncertainty.

summary(<multiScaleR_bayes>) print(<multiScaleR_bayes>) multiScale_bayes()
Estimate Scales of Effect with Bayesian Inference
check_bayes_setup() print(<multiScaleR_bayes_setup>)
Check Optional Software for Bayesian Scale Models
simulate_bayes_prior()
Simulate Prior Predictive Responses and Landscape Scales
posterior_scales()
Summarize Posterior Landscape Scales
diagnose_bayes_scales()
Check Sampling Precision of Fitted Bayesian Scales
diagnose_bayes_probabilities()
Check Monte Carlo Precision of Scale Probabilities
diagnose_bayes_laplace()
Assess Fitted Scale Probabilities from an Uncorrected Laplace Fit
plot(<multiScaleR_bayes>)
Plot Bayesian Landscape Scales and Group Variation
simulate(<multiScaleR_bayes>)
Simulate Posterior Predictive Responses

Summarize and compare models

summary(<multiScaleR>)
Summarize a multiScaleR fit
kernel_dist()
Convert Sigma into an Effective Distance
aic_tab()
Rank models by AIC or AICc
bic_tab()
Rank models by BIC
check_bayes_setup() print(<multiScaleR_bayes_setup>)
Check Optional Software for Bayesian Scale Models
cross_validate() print(<multiScaleR_cv>)
Cross-Validate Estimated Scales with Explicit Folds
cross_validate_clogit() print(<multiScaleR_choice_cv>)
Cross-Validate Conditional Choices in Complete Matched Sets
estimate_multiscale_ram() print(<multiScaleR_ram>)
Estimate Memory Requirements for Multiscale Optimization
grouped_cv_folds() print(<multiScaleR_folds>)
Create Grouped Cross-Validation Folds
msr_vars() kernel_var() landscape_var() surface_var() print(<multiScaleR_vars>)
Define Model Covariates from Raster Layers
summary(<multiScaleR_bayes>) print(<multiScaleR_bayes>) multiScale_bayes()
Estimate Scales of Effect with Bayesian Inference
prepare_gdr_histories() print(<msr_gdr_histories>)
Convert Linked Bird Records into Paired Distance-Removal Counts
prepare_gdr_intervals() print(<msr_gdr_intervals>)
Add Interval-Level Covariates to a Distance-Removal Frame
prepare_prediction() predict(<multiScaleR_prediction>) print(<multiScaleR_prediction>)
Prepare a fixed set of predictions for repeated use
print(<multiScaleR>)
Print a multiScaleR fit
print(<multiScaleR_data>)
Print prepared multiScaleR inputs
print(<summary_multiScaleR>)
Print a summary of a multiScaleR fit
sim_dat_gdr() print(<msr_gdr_simulation>)
Simulate Paired Distance-Removal Surveys with Retained Truth
validate_unmarked_data() print(<msr_unmarked_audit>)
Audit Survey Identities and Paired Distance-Removal Counts

Cross-validation

cross_validate() print(<multiScaleR_cv>)
Cross-Validate Estimated Scales with Explicit Folds
cross_validate_glm()
Cross-Validate an Ordinary GLM Reference Model
cross_validate_clogit() print(<multiScaleR_choice_cv>)
Cross-Validate Conditional Choices in Complete Matched Sets
cross_validate_field()
Cross-Validate a Spatial Response Field
cross_validate_mixed()
Cross-Validate Mixed-Effects Models
cross_validate_bayes()
Cross-Validate Bayesian Scale Models
cv_tab()
Compare Cross-Validated Prediction Scores
select_cv()
Validate Training-Only Model Selection Within CV Folds
grouped_cv_folds() print(<multiScaleR_folds>)
Create Grouped Cross-Validation Folds
spatial_cv_folds()
Create Spatially Blocked Cross-Validation Folds

Prediction and spatial projection

Apply fitted scales to new observations and to rasters.

kernel_scale.raster()
Apply Kernels to Raster Layers for Mapping
prepare_prediction() predict(<multiScaleR_prediction>) print(<multiScaleR_prediction>)
Prepare a fixed set of predictions for repeated use
predict_scales()
Predict Landscape Scales for New Covariates and Groups
predict(<multiScaleR>)
Predict from a fitted multiscale model
predict(<multiScaleR_bayes>)
Predict Means from a Bayesian Scale Model
prediction_contract
Read prediction contracts

Distance-removal models

Data preparation, fitting helpers, and audits for generalized distance-removal (GDR) models, including spatial response fields.

prepare_gdr_histories() print(<msr_gdr_histories>)
Convert Linked Bird Records into Paired Distance-Removal Counts
prepare_gdr_intervals() print(<msr_gdr_intervals>)
Add Interval-Level Covariates to a Distance-Removal Frame
expand_gdr_intervals()
Expand Site and Survey Covariates into GDR Intervals
gdr_start_values()
List the Starting-Value Order of a Distance-Removal Model
make_gdr_refitter()
Create an Audited Refit Function for Distance-Removal Models
fit_gdr_response_field()
Fit a Fixed-Range Spatial Abundance Field to a Distance-Removal Model
response_field_geometry()
Bind Projected Coordinates to a Fitted Response Field
confint(<unmarkedFitGDRField>) update(<unmarkedFitGDRField>) simulate(<unmarkedFitGDRField>) profile(<unmarkedFitGDRField>) ranef(<unmarkedFitGDRField>)
Fitted Distance-Removal Model with a Fixed-Range Abundance Field
predict(<unmarkedFitGDRField>)
Predict from a Distance-Removal Model with a Fixed-Range Abundance Field
vcov(<unmarkedFitGDRField>)
Covariance Matrix for a Distance-Removal Model with a Fixed-Range Field

Plot

plot(<multiScaleR>)
Plot method for multiScaleR objects
plot(<sigma_profile>)
Plot a Sigma Profile
plot_kernel()
Plot a Kernel Weight Curve
plot_marginal_effects()
Plot marginal effects from a fitted model

Simulation

sim_rast()
Simulate Spatially Autocorrelated Raster Surfaces
sim_dat()
Simulate Responses at Known Scales of Effect
sim_dat_unmarked()
Simulate data for optimizing scales of effect with unmarked
sim_dat_gdr() print(<msr_gdr_simulation>)
Simulate Paired Distance-Removal Surveys with Retained Truth

Example data

count_data
Simulated counts for the hab example
landscape_counts
Simulated counts for the vignette example
hab
Example habitat raster
landscape
Example landscape raster
pts
Survey points for the hab example
surv_pts
Survey points for the vignette example