Skip to content

mcp 0.3.0
Compare
Choose a tag to compare
@lindeloev lindeloev released this 03 Aug 14:24
v0.3.0
517a1ce
This commit was created on GitHub.com and signed with GitHub’s verified signature. The key has expired.
GPG key ID: 4AEE18F83AFDEB23
Expired
Learn about vigilant mode.

New features:

  • Get fits and predictions for in-sample and out-of-sample data. Read more in the article on these functions.

    • Use predict(fit) to get predicted values and quantiles.
    • Use fitted(fit) to get estimated values and quantiles.
    • Use residuals(fit) to get residuals and quantiles.

    All of the above functions include many arguments that align with (and extends) the options already in plot.mcpfit(), including getting fits/predictions for sigma (which_y = "sigma"), for the prior (prior = TRUE), and arbitrary quantiles (probs = c(0.1, 0.5, 0.999)). Use the newdata argument to get out-of-sample fitted/predicted values. Set summary = FALSE to get per-draw values.

  • Added support for weighted regression for gaussian families: model = list(y | weights(weight_column) ~ 1 + x). Weights are visualized as dot sizes in plot(fit).

  • Support for more link functions across families (e.g., family = gaussian(link = "log")) see this article for more info:

    • gaussian: "identity", "log"
    • binomial: "logit", "probit", "identity"
    • bernoulli: "logit", "probit", "identity"
    • poisson: "log", "identity"

  • New argument scale in fitted(), plot(), and fit$simulate(). When scale = "response" (default), they return fits on the observed scale. When scale = "linear", they return fits on the parameter scale where the linear trends are. Useful for model understanding and debugging.

  • Use pp_check(fit) to do prior/posterior predictive checking. See pp_check(fit, type = "x") for a list of plot types. pp_check(fit, facet_by = "varying_column") facets by a data column.

  • Improvements to plot():

    • Change point densities are now computed on a per-panel basis in plot(fit, facet_by = "varying_column"). Previous releases only displayed population-level change points.
    • You can now plot varying effects with rate = FALSE for binomial models.
    • Change point densities in plot(fit) are not located directly on the x-axis. They were "floating" 5% above the x-axis in the previous releases.

  • New argument nsamples reduces the number of samples used in most functions to speed up processing. nsamples = NULL uses all samples for maximum accuracy.

  • New argument arma in many functions toggles whether autoregressive effects should be modelled.

  • Although the API is still in alpha, feel free to try extracting samples using mcp:::tidy_samples(fit). This is useful for further processing using tidybayes, bayesplot, etc. and is used extensively internally in mcp. One useful feature is computing absolute values for varying change points: mcp:::tidy_samples(fit, population = FALSE, absolute = TRUE). Feedback is appreciated before tidy_samples will to become part of the mcp API in a future release.

Other changes

  • Change point densities in plot(fit) are now scaled to 20% of the plot for each chain X changepoint combo. This addresses a common problem where a wide posterior was almost invisibly low when a narrow posterior was present. This means that heights should only be compared within each chain x changepoint combo - not across.
  • Removed the implicit ceiling of 1000 lines and samples in plot.mcpfit().
  • Rownames are removed from ranef() and fixef() returns.
  • A major effort has been put into making mcp robust and agile to develop. mcp now use defensive programming with helpful error messages. The Test suite includes 3600+ tests.
  • plot(), predict(), etc. are now considerably faster for AR(N) due to vectorization of the underlying code.

Bug fixes

  • Sigma is now forced to stay positive via a floor at 0.
  • Fixed: support and require dplyr 1.0.0. Now also requires tidybayes 2.0.3.
  • Fixed: Parallel sampling sometimes produced identical chains.
  • Fixed several small bugs
Assets 2
Loading