NRMSampling 0.2.1

Bug fix

• NAMESPACE: Removed sample, runif, and setNames from importFrom(stats, ...). These are base:: functions and are always available without import. Listing them under stats:: caused the warning: object 'sample' is not exported by 'namespace:stats' on devtools::load_all() and R CMD CHECK.

• Added correct @importFrom stats coef lm qt sd var and @importFrom utils head read.csv roxygen2 tags to R/zzz.R so that devtools::document() regenerates the NAMESPACE without the base:: / stats:: confusion.

NRMSampling 0.2.0

New features

• Added NRMSampling.Rproj RStudio project file for streamlined development with devtools, roxygen2, and testthat integration.
• stratified_sample() now accepts a named integer vector for n_per_stratum, enabling Neyman (optimal) or any user-defined allocation.
• pps_sample() now attaches .inclusion_prob (n * p_i) to output, ready for direct use with ht_estimator().
• plot_summary() now handles all-numeric data frames when vars = NULL without requiring explicit column names.
• carbon_stock_estimate() now validates carbon_fraction range and documents the IPCC (2006) default of 0.47.
• Added sampling_efficiency() for quantitative comparison of two sampling designs via relative efficiency (RE = Var1 / Var2).
• All spatial functions now provide informative installation messages if sf or terra are not available, rather than an opaque error.

Bug fixes

• purposive_sample(): condition evaluation now uses with() instead of eval(parse(...)) on the global environment, preventing accidental access to objects outside the data frame.
• systematic_sample(): fixed an edge case where k = nrow(data) would select only one unit rather than erroring.
• quota_sample(): now silently caps quota to stratum size instead of erroring when quota exceeds available units.
• regression_estimator(): now requires at least 3 complete observations before fitting and raises an informative error.

Documentation

• .Rbuildignore updated to exclude NRMSampling.Rproj and .Rproj.user/ from the CRAN source bundle.
• .gitignore expanded with comprehensive R, RStudio, and OS entries.
• All roxygen2 @examples verified to run in under 5 seconds.
• cran-comments.md updated to reflect v0.2.0 test environments.

NRMSampling 0.1.0

Initial CRAN release

Probability Sampling

• srs_sample() – Simple random sampling with/without replacement.
• stratified_sample() – Equal, proportional, or Neyman allocation.
• systematic_sample() – Fixed-interval systematic sampling.
• cluster_sample() – Single-stage cluster sampling.
• pps_sample() – Probability-proportional-to-size sampling.

Non-Probability Sampling

• convenience_sample() – First-n convenience sample.
• purposive_sample() – Condition-based purposive sampling.
• quota_sample() – Fixed-quota non-random selection.

Estimation

• estimate_mean(), estimate_total() – Mean and total estimators.
• estimate_variance(), estimate_se() – Variance and SE with fpc.
• estimate_ci() – t-based confidence intervals.
• ratio_estimator() – Ratio estimate of population total.
• regression_estimator() – Regression-adjusted mean estimate.
• ht_estimator(), ht_variance() – Horvitz-Thompson estimator and variance.
• stratified_estimator() – Weighted stratified mean estimator.

NRM Utilities

• biomass_estimate() – Total and mean biomass from plot data.
• soil_loss_estimate() – Total and mean soil loss estimation.
• carbon_stock_estimate() – Carbon stock from biomass.
• plot_summary() – Descriptive statistics for numeric columns.
• sampling_efficiency() – Relative efficiency of two designs.

Spatial Sampling (sf + terra)

• to_sf_points(), spatial_random_sample(), spatial_systematic_sample(), spatial_stratified_sample(), spatial_cluster_sample(), raster_stratified_sample(), raster_pps_sample(), extract_raster_values(), spatial_biomass_estimate(), plot_sampling(), plot_sampling_gg().

Datasets

• sample_nrm – 100-plot synthetic NRM field dataset.
• sample_spatial – 100-observation geo-referenced NRM dataset.