ALOGIT 2023


Capabilities of ALOGIT



ALOGIT has been in intensive use by leading-edge modellers for more than 40 years and has been developed throughout that period to meet the needs of advanced modelling. As a result, ALOGIT has a high level of reliability and numerous features and facilities that are useful for professional modelling.


ALOGIT estimates the parameters of generalised logit models. The main generalisations are

  • tree logit models (also called nested or hierarchical models) allowing the alternatives in the model to be related in less restrained ways than in simple logit models but still retaining ease of use and speed of operation
  • simple mixed logit models, implemented using the flexible ‘error components’ specification, with either linear or log-linear terms in the utility function
  • size variables, representing the quantity rather than the quality of an alternative.

ALOGIT performs four key functions.


1. DATA INPUT:

  • revealed preference or stated preference, disaggregate or aggregate data can be used (or any combination of these)
  • choice, ranking or proportional split data can be employed
  • input data can be manipulated and transformed freely to allow the user freedom in finding the required explanation of behaviour
  • extensive testing of input data to reveal modelling problems
  • simple controls, yet giving access when required to full sophistication
  • multiple data sets of different formats can be analysed simultaneously, either in succession or linked using named (key) variables
  • common data can be input and used with all observations

2. MODEL ESTIMATION:

  • all coefficients are estimated simultaneously using maximum likelihood estimation (i.e. ‘full information’ estimates)
  • models can be binomial, multinomial, or tree (nested) logit models, with unlimited branches and levels
  • alternatively, simple mixed logit or error component analysis, including differing distributions and correlated error terms
  • non-linear utility functions allow size (attraction) variables to be included correctly
  • composite alternatives can be indicated as chosen
  • exact (analytical) first and second derivatives​ are used, for speed and accuracy in convergence and better estimation of errors
  • coefficient estimates, classical and 'robust' standard errors and correlations of errors,
  • consumer surplus measures and several detailed tests are all standard, with informative, clearly labelled, well-laid-out output, suitable for immediate incorporation in reports
  • a function in the ALOGIT Shell can be used to make comparisons between the estimates for different model variants for presentation and efficient model development
  • an option for initial linear estimation reduces run time for complex models
  • problem sizes are effectively not limited by ALOGIT

3. FORECASTING:

  • the user can specify detailed scenarios which incorporate a series of changes in the variables influencing choice, and ALOGIT can predict, display and analyse the consequent changes in behaviour
  • a function in the ALOGIT Shell can then be used to make graphical and tabular presentations of scenario outputs
  • alternatively, output can be made to other programs such as Excel

4. DATA PROCESSING:

  • ALOGIT can be used for a range of simple data processing tasks, using the control language and statistical reporting procedures to give an efficient working environment

All 4 of these functions are controlled by ALOGIT’s much-improved control file, with

  • a substantially improved command language, including named variables, intuitively appealing definition structure for hierarchical models, named Boolean operators (TRUE, FALSE, AVAIL etc.)
  • include-file option for use of external files with command lines or coefficients, streamlining model (estimation) management
  • array definition of alternatives, system data items and variables
  • random number generator (using uniform, normal, logistic distributions or assignment of multinomial variable with specific probabilities)
  • ‘if ... THEN ... ELSE... END’ and ‘DO ... END’ syntax, to simplify data transformations
  • intuitive specification of tree logit models using $NEST commands.