Column names: tidy()

[alexpghayes]

This issue thread is for proposals for column names changes. Please read #452 before commenting.

[alexpghayes]

Existing column names (this is fairly complete, but not perfect -- it was scraped from the \value{} section of .Rd files)

• column: '...' description: The initial columns correspond to the grouping factors on the right hand side of the model formula. used_by: survdiff
• column: cumulative description: Cumulative percentage of variation explained. used_by:
  • tidy_irlba
  • tidy_svd
  • prcomp
• column: PC description: An integer vector indicating the principal component. used_by:
  • tidy_irlba
  • tidy_svd
  • prcomp
• column: percent description: Percentage of variation explained. used_by:
  • tidy_irlba
  • tidy_svd
  • prcomp
• column: row description: Needs custom description used_by:
  • tidy_irlba
  • tidy_svd
  • prcomp
• column: std.dev description: Standard deviation explained by this PC. used_by:
  • tidy_irlba
  • tidy_svd
  • prcomp
• column: value description: Needs custom desciption. used_by:
  • tidy_irlba
  • tidy_svd
  • prcomp
• column: acf description: calculated correlation used_by: acf
• column: adj.p.value description: P-value adjusted for multiple comparisons used_by: TukeyHSD
• column: alternative description: Alternative hypothesis (character). used_by:
  • binWidth
  • htest
• column: bias description: Bias of the statistic. used_by: boot
• column: ci.width description: Expected width of confidence interval. used_by: binWidth
• column: class description: The class under consideration. used_by:
  • confusionMatrix
  • poLCA
• column: cluster description: A factor describing the cluster from 1:k used_by: kmeans
• column: column1 description: Name or index of the first column being described used_by: rcorr
• column: column2 description: Name or index of the second column being described used_by: rcorr
• column: comparison description: Levels being compared, separated by - used_by: TukeyHSD
• column: component description: Cluster id as a factor. For a model k clusters, these will be as.factor(1:k), or as.factor(0:k) if there's a noise term. used_by: Mclust
• column: conf.high description: The upper end of a confidence interval for the term under consideration. Included only if conf.int = TRUE. used_by:
  • Arima
  • betareg
  • biglm
  • btergm
  • cch
  • confusionMatrix
  • cv.glmnet
  • felm
  • glm
  • glmRob
  • gmm
  • htest
  • ivreg
  • kappa
  • lavaan
  • lm
  • lmodel2
  • lmRob
  • mjoint
  • nlrq
  • nls
  • plm
  • rlm
  • rq
  • rqs
  • speedlm
  • survfit
  • survreg
  • TukeyHSD
• column: conf.low description: The lower end of a confidence interval for the term under consideration. Included only if conf.int = TRUE. used_by:
  • Arima
  • betareg
  • biglm
  • btergm
  • cch
  • confusionMatrix
  • cv.glmnet
  • felm
  • glm
  • glmRob
  • gmm
  • htest
  • ivreg
  • kappa
  • lavaan
  • lm
  • lmodel2
  • lmRob
  • mjoint
  • nlrq
  • nls
  • plm
  • rlm
  • rq
  • rqs
  • speedlm
  • survfit
  • survreg
  • TukeyHSD
• column: cutoff description: The cutoff used for classification. Observations with predicted probabilities above this value were assigned class 1, and observations with predicted probabilities below this value were assigned class 0. used_by: roc
• column: denominator description: The denominator, which is tau=kendall_score/denominator. used_by: Kendall
• column: dev.ratio description: fraction of null deviance explained at each value of lambda used_by: glmnet
• column: df description: Degrees of freedom used by this term in the model. used_by:
  • anova
  • aov
  • aovlist
• column: distance description: Distance between items used_by: dist
• column: estimate description: Estimated value of the term under consideration. Needs additional custom info. used_by:
  • aareg
  • Arima
  • betareg
  • biglm
  • btergm
  • cch
  • coeftest
  • confusionMatrix
  • coxph
  • cv.glmnet
  • ergm
  • felm
  • fitdistr
  • gamlss
  • garch
  • glm
  • glmnet
  • glmRob
  • gmm
  • htest
  • ivreg
  • kappa
  • lavaan
  • lm
  • lmodel2
  • lmRob
  • mjoint
  • muhaz
  • multinom
  • nlrq
  • nls
  • plm
  • poLCA
  • rcorr
  • ridgelm
  • rlm
  • rq
  • rqs
  • speedlm
  • survexp
  • survfit
  • survreg
  • TukeyHSD
• column: estimate1 description: Sometimes two estimates are computed, such as in a two-sample t-test used_by: htest
• column: estimate2 description: Sometimes two estimates are computed, such as in a two-sample t-test used_by: htest
• column: event description: observed number of events used_by: pyears
• column: exp description: weighted expected number of events in each group used_by: survdiff
• column: expected description: expected number of events (present only if a ratetable term is present) If the data.frame = TRUE argument is supplied to pyears, this is simply the contents of x$data. used_by: pyears
• column: flX description: Factor loading of term on factor X. There will be as many columns of this format as there were factors fitted. used_by: factanal
• column: fpr description: The false positive rate at the given cutoff. used_by: roc
• column: GCV description: generalized cross validation value for this lambda used_by: ridgelm
• column: group description: The group (if specified) in the lavaan model used_by: lavaan
• column: group1 description: First group being compared used_by: pairwise.htest
• column: group2 description: Second group being compared used_by: pairwise.htest
• column: index description: Index (i.e. date or time) for a ts or zoo object. used_by:
  • ts
  • zoo
• column: item1 description: First item used_by: dist
• column: item2 description: Second item used_by: dist
• column: kendall_score description: Kendall score. used_by: Kendall
• column: lag description: lag values used_by: acf
• column: lambda description: Value of penalty parameter lambda. used_by:
  • cv.glmnet
  • glmnet
  • ridgelm
• column: mcmc.error description: The MCMC error used_by: ergm
• column: mean description: The mean for each component. In case of 2+ dimensional models, a column with the mean is added for each dimension. NA for noise component used_by: Mclust
• column: meansq description: Mean sum of squares. Equal to total sum of squares divided by degrees of freedom. used_by:
  • anova
  • aov
  • aovlist
• column: method description: Needs custom info. used_by:
  • htest
  • lmodel2
• column: 'n' description: Needs custom info. used_by:
  • binDesign
  • binWidth
  • pyears
  • rcorr
• column: 'N' description: number of subjects in each group used_by: survdiff
• column: n.censor description: number of censored events used_by: survfit
• column: n.event description: number of events at time t used_by: survfit
• column: n.risk description: Number of individuals at risk at time zero. used_by:
  • survexp
  • survfit
• column: num.df description: Degrees of freedom used_by: manova
• column: nzero description: Number of coefficients that are exactly zero for given lambda. used_by: cv.glmnet
• column: obs description: weighted observed number of events in each group used_by: survdiff
• column: op description: The operator in the model syntax (e.g. ~~ for covariances, or ~ for regression parameters) used_by: lavaan
• column: outcome description: Outcome of manifest variable used_by: poLCA
• column: p description: True proportion. used_by: binWidth
• column: p.value description: Needs custom information. used_by:
  • aareg
  • anova
  • aov
  • aovlist
  • betareg
  • biglm
  • cch
  • coeftest
  • confusionMatrix
  • coxph
  • ergm
  • felm
  • gamlss
  • garch
  • glm
  • glmRob
  • gmm
  • htest
  • ivreg
  • Kendall
  • lavaan
  • lm
  • lmRob
  • manova
  • mjoint
  • multinom
  • nlrq
  • nls
  • pairwise.htest
  • plm
  • rcorr
  • rlm
  • rq
  • rqs
  • speedlm
  • survreg
• column: parameter description: Needs custom info. used_by:
  • tidy_optim
  • gamlss
  • htest
• column: power description: Power achieved for given value of n. used_by: binDesign
• column: proportion description: The mixing proportion of each component used_by: Mclust
• column: pyears description: person-years of exposure used_by: pyears
• column: response description: Which response column the coefficients correspond to (typically Y1, Y2, etc). used_by:
  • glm
  • glmRob
  • lm
  • lmRob
  • rlm
  • speedlm
• column: robust.se description: robust version of standard error estimate (only when x was called with dfbeta = TRUE) used_by: aareg
• column: scale description: Scaling factor of estimated coefficient used_by: ridgelm
• column: series description: Name of the series (present only for multivariate time series). used_by:
  • ts
  • zoo
• column: size description: Number of points assigned to cluster. used_by:
  • kmeans
  • Mclust
• column: state description: state if multistate survfit object inputted used_by: survfit
• column: statistic description: Needs custom information used_by:
  • aareg
  • anova
  • aov
  • aovlist
  • betareg
  • biglm
  • boot
  • cch
  • coeftest
  • coxph
  • felm
  • gamlss
  • garch
  • glm
  • glmRob
  • gmm
  • htest
  • ivreg
  • Kendall
  • lavaan
  • lm
  • lmRob
  • manova
  • mjoint
  • multinom
  • nlrq
  • nls
  • plm
  • rlm
  • rq
  • rqs
  • speedlm
  • survreg
• column: std.all description: Standardized estimates based on both the variances of both (continuous) observed and latent variables. used_by: lavaan
• column: std.error description: Standard error of the estimate. boot needs custom info. used_by:
  • aareg
  • Arima
  • betareg
  • biglm
  • boot
  • cch
  • coeftest
  • coxph
  • cv.glmnet
  • ergm
  • felm
  • fitdistr
  • gamlss
  • garch
  • glm
  • glmRob
  • gmm
  • ivreg
  • lavaan
  • lm
  • lmRob
  • mjoint
  • multinom
  • nlrq
  • nls
  • plm
  • poLCA
  • rlm
  • rq
  • rqs
  • speedlm
  • survfit
  • survreg
• column: std.lv description: Standardized estimates based on the variances of the (continuous) latent variables only used_by: lavaan
• column: std.nox description: Standardized estimates based on both the variances of both (continuous) observed and latent variables, but not the variances of exogenous covariates. used_by: lavaan
• column: step description: which step of lambda choices was used used_by: glmnet
• column: strata description: strata if stratified survfit object inputted used_by: survfit
• column: stratum description: The error stratum used_by: aovlist
• column: sumsq description: Sum of squares explained by this term used_by:
  • anova
  • aov
  • aovlist
• column: term description: Needs custom info. used_by:
  • aareg
  • anova
  • aov
  • aovlist
  • Arima
  • betareg
  • biglm
  • boot
  • btergm
  • cch
  • coeftest
  • confusionMatrix
  • ergm
  • felm
  • fitdistr
  • gamlss
  • garch
  • glm
  • glmnet
  • glmRob
  • gmm
  • ivreg
  • lavaan
  • lm
  • lmodel2
  • lmRob
  • mjoint
  • multinom
  • nlrq
  • nls
  • plm
  • ridgelm
  • rlm
  • rq
  • rqs
  • speedlm
  • survreg
  • TukeyHSD
• column: time description: Point in time. used_by:
  • muhaz
  • survexp
  • survfit
• column: tpr description: The true positive rate at the given cutoff. used_by: roc
• column: type description: Either "weighted" or "unweighted" used_by: kappa
• column: uniqueness description: Proportion of residual, or unexplained variance used_by: factanal
• column: value description: Needs custom info. used_by:
  • tidy_optim
  • ts
  • zoo
• column: var_kendall_score description: Variance of the kendall_score. used_by: Kendall
• column: variable description: Needs custom info. used_by:
  • factanal
  • poLCA
• column: variance description: In case of one-dimensional and spherical models, the variance for each component, omitted otherwise. NA for noise component used_by: Mclust
• column: withinss description: The within-cluster sum of squares used_by: kmeans
• column: y.value description: The response level used_by: multinom
• column: z description: z score used_by: aareg

[alexpghayes]

Some current conflicts I am aware of for tidy() (some of these not in the list above):

• acf / autocorrelation
• comp / component
• conf.high / upper
• conf.low / lower
• exp / expected
• df / num.df
• n / N / size
• parameter / term / variable
• p / proportion
• strata / stratum
• response / y.level / y.value
• statistic / z

It seems like there's also some redundancy in column names that describe "percent of variance" explained type information.

[copernican]

nzero is used by cv.glmnet(). The description above is "number of coefficients that are exactly zero for given lambda." I think this may be reversed. The documentation for cv.glmnet() states that this is actually the "number of non-zero coefficients at each lambda." If you agree (and if appropriate), I am happy to file an issue in modeltests instead, or to submit a PR.

[alexpghayes]

A PR or an issue would be ideal!

[github-actions[bot]]

This issue has been automatically closed due to inactivity.

[github-actions[bot]]

This issue has been automatically locked. If you believe you have found a related problem, please file a new issue (with a reprex: https://reprex.tidyverse.org) and link to this issue.