LoginSign Up
← Back to all sparks
R

RandomWalker

ANALYTICS
Velocity0.0

R package RandomWalker by spsanderson — release notes from GitHub.

github.com/spsanderson/RandomWalker ↗

A random-walk generator that outgrew one dimension and renamed its core column to prove it.

stochastic-processesrandom-walkssimulationtidyverser-package
◆Current state
RandomWalker generates families of stochastic paths — Brownian motion, geometric Brownian motion, drift walks, discrete walks — as tidy tibbles, with cumulative-statistic augmenters, summarisers and a visualize_walks() plotting layer on top. The development series before 1.0.0 extended generation to two and three dimensions and renamed the step index from x to step_number, which is the shape the package now carries into its stable release.
◆Where it's heading
The package built outward in clear stages: generators first, then a set of std_cum_*_augment() transformations over the results, then the dimensional generalisation that forced the column rename. That progression suggests a design settling on walks as a tidy data structure to be transformed and plotted rather than a set of one-off simulators. The 1.0.0 tag itself carries no release notes in this feed — its body is stray YAML front matter — so the milestone's own contents cannot be read here.
◆Prediction
With dimensions generalised and a 1.0.0 cut, further work most plausibly extends the augmenter and summariser layer to multi-dimensional walks. The empty 1.0.0 body means any specific claim about what the stable release contains would be guesswork.

◆Recent moves

  1. 0y ago

    RandomWalker 1.0.0

    The 1.0.0 stable tag, arriving after the dimensional generalisation and column rename landed in the development series. The feed body contains only YAML front matter, so what this release adds beyond declaring stability is not readable here.

    View source ↗
  2. 1y ago

    Random walks gain up to three dimensions

    ⚡ SPARK

    Random walks gain a .dimensions parameter supporting up to three dimensions, which forced the old x column to become step_number across every generator including rw30(). This is the release that turned a one-dimensional simulator into a general path generator, and it is where the package's current data shape was set.

    View source ↗
  3. 1y ago

    Cumulative-statistic augmenters and interactive plotting

    Seven new functions land at once — five std_cum_*_augment() variants for cumulative sum, product, min, max and mean, plus get_attributes() and running_quantile(). visualize_walks() also gains .interactive and .pluck parameters, establishing the transform-and-plot layer over generated walks.

    View source ↗
  4. 1y ago

    Initial release: six walk generators plus visualisation

    The initial release, shipping the generator set — brownian_motion(), geometric_brownian_motion(), random_normal_walk(), random_normal_drift_walk(), discrete_walk() and rw30() — alongside euclidean_distance(), visualize_walks() and summarize_walks(). The surface everything since has extended.

    View source ↗