tidyr
tidyr replaced separate() with a family that says what it does.
Comparison · Analytics
performance vs statsmodels
A side-by-side editorial comparison of performance and statsmodels — release velocity, themes, recent moves, and the top alternatives to consider.
performance vs statsmodels: at a glance
What is performance?
performance keeps adding ways to check a model you have already fitted.
performance is at 0.17.1, which added check_priors() for prior predictive checks on Bayesian models and gave check_overdispersion(), check_model() and check_predictions() arguments to control residual type and plot range. The releases before it are a similar mix: a -2LL criterion column in test_likelihoodratio(), Bayesian predictive checks routed through modelbased, and in 0.16.0 a set of breaking renames including RMSA to the correct RMSR.
What is statsmodels?
statsmodels ships only what the ecosystem breaks — six releases, no new statistics.
Every release in this window is a compatibility release. 0.14.2 and 0.14.3 absorbed NumPy 2, 0.14.5 fixed an import failure caused by SciPy 1.16, and 0.14.6 did the same for pandas 3.0. The only additive change across two years is Pyodide support in 0.14.4, described in its own notes as one feature and no fixes. A 0.15.0.dev0 tag exists from 2023 and has not been followed by a 0.15 release.
performance vs statsmodels: editorial side-by-side
performance
0.0ANALYTICS
performance keeps adding ways to check a model you have already fitted.
◆ Current state
performance is at 0.17.1, which added check_priors() for prior predictive checks on Bayesian models and gave check_overdispersion(), check_model() and check_predictions() arguments to control residual type and plot range. The releases before it are a similar mix: a -2LL criterion column in test_likelihoodratio(), Bayesian predictive checks routed through modelbased, and in 0.16.0 a set of breaking renames including RMSA to the correct RMSR.
◆ Where it's heading
Two consistent habits. Diagnostics keep gaining arguments to narrow what is examined — ppc_range, x_limits, maximum_dots, show_ci — which reads as a package being used on models large and awkward enough that the defaults stopped working. And simulated residuals via DHARMa keep displacing standard ones as the basis for the checks themselves.
◆ Prediction
With check_priors() newly added and Bayesian predictive checks now routed through modelbased, the next release most likely extends the Bayesian diagnostic set rather than reworking the frequentist checks.
statsmodels
0.0ANALYTICS
statsmodels ships only what the ecosystem breaks — six releases, no new statistics.
◆ Current state
Every release in this window is a compatibility release. 0.14.2 and 0.14.3 absorbed NumPy 2, 0.14.5 fixed an import failure caused by SciPy 1.16, and 0.14.6 did the same for pandas 3.0. The only additive change across two years is Pyodide support in 0.14.4, described in its own notes as one feature and no fixes. A 0.15.0.dev0 tag exists from 2023 and has not been followed by a 0.15 release.
◆ Where it's heading
The library is being kept alive rather than developed: each release answers a break introduced upstream, and the interval between them is set by the NumPy, SciPy and pandas release calendars rather than by anything statsmodels is building. Two consecutive releases whose stated purpose was restoring the ability to import the package is the sharpest available signal about maintainer bandwidth. The 0.15 line remains a dev tag with no visible progress toward a release.
◆ Prediction
The next release is most likely another compatibility patch triggered by a NumPy, SciPy or pandas major, and nothing in these entries indicates 0.15 is close.
Alternatives to performance and statsmodels
Other Analytics products tracked by Sparkpulse, ranked by recent ship velocity. Each card links to a full editorial trajectory and lets you pivot into a head-to-head comparison with either performance or statsmodels.
modeltime
modeltime built conformal intervals in, then went quiet on features.
cmdstanpy
CmdStanPy is clearing deprecations ahead of a 2.0 it keeps announcing.
dowhy
DoWhy adds one estimation method a year and keeps its identification edge.
OpenHouse
OpenHouse is hardening the seams where table policies and jobs quietly fail.
silx
silx 3.0 moved its default Qt binding to PySide6 — a migration for everyone embedding it.
See all performance alternatives → · See all statsmodels alternatives →
Recent activity from performance and statsmodels
Latest ship moves from both products, interleaved chronologically. ⚡ = editorial spark.
- 1mo agoperformancecheck_priors() added; overdispersion plots use simulated residuals
- 2mo agoperformance-2LL criterion column and unified Bayesian predictive checks
- 6mo agoperformanceBreaking renames plus point-count and CI controls in check_model()
- 8mo agostatsmodels0.14.6: restores importing under pandas 3.0
- 8mo agoperformancecheck_autocorrelation() methods for DHARMa objects
- 10mo agoperformanceFixes CRAN checks after an rstanarm update
- 11mo agoperformancetinytable output format in display()
- 1y agostatsmodels0.14.5: restores importing under SciPy 1.16
- 1y agostatsmodels0.14.4: Pyodide support
- 1y agostatsmodels0.14.3: NumPy 2 environments and corrected macOS builds
- 2y agostatsmodels0.14.2: full NumPy 2 compatibility
- 2y agostatsmodelsRelease 0.14.1
Frequently asked questions
What is the difference between performance and statsmodels?
They serve adjacent needs but don't currently overlap on shipped themes. performance and statsmodels are shipping at a similar cadence (velocity 0.0 vs 0.0, both within Sparkpulse's "active" band). See the at-a-glance table above for a side-by-side breakdown of velocity, recent sparks, and editorial themes.
Is performance better than statsmodels?
Sparkpulse doesn't pick a winner — we score release velocity, not feature parity. performance and statsmodels are shipping at a similar cadence (velocity 0.0 vs 0.0, both within Sparkpulse's "active" band). For your specific use case, the alternatives sections above list other Analytics products to evaluate alongside.
What are the best alternatives to performance?
Top performance alternatives in Analytics are ranked by recent ship velocity. Browse the "performance alternatives" section above for the current picks, or visit /alternatives/easystats-performance for the full list with editorial commentary on each.
What are the best alternatives to statsmodels?
Top statsmodels alternatives in Analytics are ranked by recent ship velocity. Browse the "statsmodels alternatives" section above for the current picks, or visit /alternatives/statsmodels for the full list with editorial commentary on each.