What’s New in LIMDEP Version 11 Software? LIMDEP Version 11 continues the expansion of our premier software for cross section, panel data and time series analysis. Version 11 contains major new extensions to the program for estimation and statistical analysis of econometric models and a long list of new models and features. New features include:
Stochastic Frontier Models
- New parametric specifications
- Normal-Rayleigh
- New FIML estimator for sample selection correction
- Sample selection with panel data
- Partially inefficient latent class stochastic frontier model
Panel Data Stochastic Frontier Models
- Rayleigh model
- Belotti and Ilardi partial differences
- True fixed and random effects
- Wikstrom method of moments
- Generalized true random effects
New Partial Effects and Simulation Tools
- New graphical devices
- Transition matrices for categorical variables
- Krinsky and Robb method for standard errors
- Extended to more models
Expanded Graphics
Histograms
- Overlay histograms with box plots or kernel density estimators
- Group up to 4 histograms
- Up to 40 plots in one figure
- New BOXPLOTS command
- Simultaneous plots
Kernel Density Estimator
- Several new specifications for plotting kernel density estimators
- Simultaneous plots
Treatment Effects Procedures
- Endogenous treatment effects
- FIML estimation for probit, ordered probit, Poisson and negative binomial
New Tools for Estimation and Inference
New Analysis Tools
- Over 30 new CREATE functions for panel data transformations
- Dozens of new functions for CALCULATE and MATRIX
- New random number generators
- Cluster covariance matrices for all models
- Inverse hyperbolic sine model
- Fractional regression for cross section and panel data
- New panel data functions for MAXIMIZE
- Obtain separate strata means for stratified data
- New commands for correlation matrices
- New TABLE command for descriptive statistics by strata
- New commands for quantiles and frequencies
- New Cronbach’s alpha correlation device
- Observation labels
- Labellists
- Observation tags
- Generalized beta of the second kind
- Generalized gamma model
- New specifications for bootstrap and jackknife iterations
- Block bootstrap for panel data
What’s New in NLOGIT Version 6 Software? All the new features described for LIMDEP 11 are in NLOGIT 6. In addition, there are many new features in Version 6. We have added several enhancements to give you greater flexibility in analyzing different types of data. Many of the features of NLOGIT, existing and new, are designed to let you go beyond just computing coefficients, to analyzing and using your model. We have added many new models including the random regret logit model and best/worst outcome. NLOGIT 6 continues to pioneer new developments for estimation in WTP (willingness to pay) space. Altogether, we have added dozens of features in NLOGIT 6, some clearly visible ones such as the new models and some ‘behind the scenes’ that will smooth the operation and help to stabilize the estimation programs. The following will summarize the important new developments.
New Multinomial Choice Models
NLOGIT 6 includes many new commands and extension of the random parameters model and latent class models:
- Fixed effects in multinomial logit models
- Random effects multinomial logit models
- Random regret logit model
- Best/worst outcome data
- Berry, Levinsohn and Pakes random parameters logit model
- Latent classes with random parameters
- Generalized mixed logit
Model Extensions
- Willingness to pay
- Attribute nonattendance (explicit and implicit)
- Individual specific expected parameters
- Model simulation
- Estimated elasticities and partial effects
- Robust covariance matrix
- Random data generators
- Posterior estimates from latent class models
- Coefficients in random parameters models
- Simplified WALD command