v0.7.7

RegressionTables v0.7.7

Diff since v0.7.6

Merged pull requests:

• Switch to broadcasting to fix no length issue on rhs in formula (#165) (@junder873)

v0.7.6

RegressionTables v0.7.6

Diff since v0.7.5

Merged pull requests:

• Force name and level to be of type string in CategoricalCoefName, not some funny string type like string7. (#162) (@grahamstark)
• no longer demand StatsModels.formula to be defined on every new model (#163) (@junder873)

Closed issues:

• Is there a way to display "+" for p value between 5% and 10%？ (#156)
• Does this handle auxilliary parameters? (#158)
• Struggling to understand documentation (#160)

v0.7.5

RegressionTables v0.7.5

Diff since v0.7.4

Merged pull requests:

• Use StatsAPI.confint instead of custom confidence interval function (#155) (@junder873)

v0.7.4

RegressionTables v0.7.4

Diff since v0.7.3

Merged pull requests:

• Fix BIC calling aaic instead of bic (#154) (@junder873)

Closed issues:

• Documentation seems to suggest conflicting argument types for regression_statistics and neither example works in practice: please reproduce? (#152)

v0.7.3

RegressionTables v0.7.3

Diff since v0.7.2

Merged pull requests:

• Add support for Typst tables (#141) (@junder873)

v0.7.2

RegressionTables v0.7.2

Diff since v0.7.1

Merged pull requests:

• Add vertical gaps function (#151) (@junder873)

v0.7.1

RegressionTables v0.7.1

Diff since v0.7.0

Merged pull requests:

• Use Format.jl instead of Formatting.jl to fix bugs (#149) (@ScottPJones)

v0.7.0

RegressionTables v0.7.0

Diff since v0.6.2

Merged pull requests:

• Change how below statistics work to make more flexible (#146) (@junder873)

Closed issues:

• bootstrap confidence intervals (#145)

Breaking Changes

This version contains a couple breaking internal changes:

• All below statistics (StdError, TStat, ConfInt) now expect the RegressionModel and which coefficient to calculate their respective values. This is instead of several other values (standard error, coefficient value and degrees of freedom). The purpose of this is to make any below statistic possible since the RegressionModel itself is passed.
• The standardize_coef_values function now operates on a single value at a time (instead of the whole vector of coefficients and standard errors). This likely has a slight performance penalty but makes the previous change possible.
• A new function can_standardize is used for each RegressionModel type to check if it is possible to standardize the type. This is by default false so the standardize_coef_values function is only necessary if can_standardize is set to true.

v0.6.2

RegressionTables v0.6.2

Diff since v0.6.1

Merged pull requests:

• Add option to use original coefficient names in arguments (#148) (@junder873)

Closed issues:

• using term instead of @formula from FixedEffectModels causes an error (#142)
• Confusion extending regtable for different type of RegressionModel (#144)
• keep option uses post-processing variable names: document or change? (#147)

v0.6.1

RegressionTables v0.6.1

Diff since v0.6.0

Merged pull requests:

• Updates related to FixedEffectModels@1.11 (#143) (@junder873)

Closed issues:

• fixed effect name bug? (#140)