Stata Blogs
- 140: Shorter or fewer category labels with graph bar [21(1)]
- 139: The by() option of graph can work better than graph combine
- 138: Local macros have local scope
- 137: Interpreting constraints on slopes of rank-deficient design matrices
- 136: Between-group comparisons in a scatterplot with weighted markers
- 135: Leaps and bounds
- 134: Multiplicative and marginal interaction effects in nonlinear models
- 133: Box plots that show median and quartiles only
- 131: Custom legends for graphs that use translucency
- Tips for calculating and displaying risk-standardized hospital outcomes in Stata
- 130: 106610 and all that: Date variables that need to be fixed
- 129: Efficiently processing textual data with Stata's new Unicode features
- 128: Marginal effects in log-transformed models: A trade application
- 127: Use capture noisily groups
- 126: Handling irregularly spaced high-frequency transactions data
- 125: Binned residual plots for assessing the fit of regression models for binary outcomes
- 124: Passing temporary variables to subprograms
- 123: Spell boundaries
- 122: Variable bar widths in two-way graphs
- 121: Box plots side by side
- 120: Certifying subroutines
- 119: Expanding datasets for graphical ends
- 118: Orthogonalizing powered and product terms using residual centering
- 117: graph combine–Combining graphs
- 116: Where did my p-values go? (Part 3)
- 115: How to properly estimate the multinomial probit model with heteroskedastic errors
- 114: Expand paired dates to pairs of dates
- 113: Changing a variable's format: What it does and does not mean
- 112: Where did my p-values go? (Part 2)
- 111: More on working with weeks (erratum)
- 110: How to get the optimal k-means cluster solution
- 109: How to combine variables with missing values
- 108: On adding and constraining
- 107: The baseline is now reported
- 106: With or without reference
- 105: Daily dates with missing days
- 104: Added text and title options
- 103: Expressing confidence with gradations
- 102: Highlighting specific bars
- 101: Previous but different
- 100: Mata and the case of the missing macros
- 99: Taking extra care with encode
- 98: Counting substrings within strings
- 97: Getting at ρ's and σ's
- 96: Cube roots
- 95: Estimation of error covariances in a linear model
- 94: Manipulation of prediction parameters for parametric survival regression models
- 93: Handling multiple y axes on twoway graphs
- 92: Manual implementation of permutations and bootstraps
- 91: Putting unabbreviated varlists into local macros
- 90: Displaying partial results
- 89: Estimating means and percentiles following multiple imputation
- 88: Efficiently evaluating elasticities with the margins command
- 87: Interpretation of interactions in nonlinear models
- 86: The missing() function
- 85: Looping over nonintegers
- 84: Summing missings
- 83: Merging multilingual datasets
- 81: A table of graphs
- 80: Constructing a group variable with specified group sizes
- 79: Optional arguments to options
- 78: Going gray gracefully: Highlighting subsets and downplaying substrates
- Evaluating concavity for production and cost functions
- 75: Setting up Stata for a presentation
- 74: firstonly, a new option for tab2
- 68: Week assumptions
- 67: J() now has greater replicating powers
- 66: ds—A hidden gem
- 62: Plotting on reversed scales
- 58: nl is not just for nonlinear models
- 56: Writing parameterized text files
- 54: Post your results
- 52: Generating composite categorical variables
- 50: Efficient use of summarize
- 46: Step we gaily, on we go
- 42: The overlay problem: Offset for clarity
- 38: Testing for groupwise heteroskedasticity
- 35: Detecting whether data have changed
- 32: Do not stop
- 29: For all times and all places
- 26: Maximizing compatibility between Macintosh and Windows
- 23: Regaining control over axis ranges
- 20: Generating histogram bin variables
- 15: Function graphs on the fly
- 14: Using value labels in expressions
- 12: Tuning the plot region aspect ratio
- 11: The nolog option with maximum–likelihood modeling commands
- 8: Splitting time–span records with categorical time–varying covariates
- 4: Using display as an online calculator
- 3: How to be assertive
- 2: Building with floors and ceilings
- 1: The eform() option of regress
No comments:
Post a Comment