Here's an overview of the features introduced in Analytica
release 3.1 added since release 3.0.
You can upgrade
free to release 3.1 if you have current maintenance
and technical support.
New
Editions
Analytica 3.1
adds some important editions to the existing product line-up: Lite,
Professional, Enterprise, and Analytica Decision Engine (ADE):
Power
Player
The Power Player lets you distribute to end users powerful applications
you have created in Analytica Enterprise. Like the standard Player,it
lets users browse models, change inputs, compute and view results.
Unlike the standard Player, it also lets users save their models
with changed inputs for continuing in future sessions, and it supports
Enterprise-level features, including OLE linking, reading and writing
relational databases via ODBC, Huge Arrays, and the available Optimizer.
See feature comparison. To order.
The
Optimizer
The Analytica Optimizer lets you find optimal decisions with your
Analytica models. It can maximize expected value, minimize costs,
or optimize any quantified objective, subject to constraints. It
uses a powerful, general purpose optimizer and equation solver.
It includes state-of-the-art algorithms for linear programming (LP)
and quadratic programming (QP) for up to 2000 decision variables
and 8000 constraints. For nonlinear programming (NLP) , it provides
hybrid methods using classical gradient-search and evolutionary
(genetic) algorithms for smooth and discontinuous objective functions,
with up to 500 decision variables and 250 constraints. It can handle
continuous, binary, integer, or mixed-integer variables. It can
also solve linear and non-linear systems of equations.
We now offer
Educational versions of all Editions of Analytica, including Lite,
Professional, Enterprise, Optimizer, Power Player, and ADE. (Previously,
we offered only Lite and Professional for Educational use.). See feature
comparison. To order see the educational
order page.
New
functions
Matrix
operations
MatrixMultiply
-- matrix multiplication, EigenDecomp Computation of Eigen
vectors and Eigen values.
Probability
distributions
Yet more ways
to express uncertainties and statistical analysis: Pert, and
ways to specify Normal, Lognormal, Beta, and others using specified
percentiles (fractiles) or mean and standard deviation instead of
more abstract parameters.
Multivariate
and correlated probability distribution
Perhaps, the
most challenging aspect of modeling uncertainty is expressing probabilistic
dependence and correlations. Analytica 3.1 makes it easier to do this
with a library of new functions, including Multinormal, Multinomial,
and Dirichlet, as well as functions for creating custom distributions
with any shape and rank-correlation matrix.
New
function libraries
New libraries
include a variety of powerful new functions, including functions for
data analysis, principal components analysis, reindexing, dynamic
models, and multiattribute decision analysis.
Extended
syntax
Name-based
syntax
for calling functions
For user-defined functions and some built-in functions, you can
use name-based syntax, as an alternative to conventional position-based
syntax in specifying parameters — e.g. Gaussian(meanVec: M,
covar: C, i: rows, j: cols)
This is clearer, allows nontrailing parameters to be optional,
and is similar to Analytica's name-based subscripting — e.g.
Sales[Time = 2005, Product = 'Optimizer']
Parameter
qualifiers
We introduce
a much richer set of qualifiers for parameter declarations to what
specify what type of parameter is expected and how it should be treated.
These include: Evaluation Mode Qualifiers controlling whether
a parameter should be evaluated deterministically (Determtype), probabilistically
(ProbType) or passed as a Variable name (VarType) or Index (IndexType).
Type Checking Qualifiers include Numeric, Positive, TextType,
and ReferenceType. Array Type Qualifiers to make sure functions
generalize appropriately for array-valued parameters. Any parameter
can be optional, using the name-based calling syntax.
Assignment
to globals
You can now assign to global
variables from scripts in Buttons, allowing more powerful interactive
functions, still avoiding dangerous side effects.
Convenience
Keyboard
shortcuts
The F2 key lets you switch rapidly between edit and
arrow modes. When typing a Definition into the Attribute pane, click
on another node in this diagram while pressing ALT
key to insert its name into the Definition.
Graphing
discrete
distributions
Analytica makes
better guesses about whether a computed distribution is discrete,
graphing it as a probability mass with solid bars, versus using histogram
style for probability density functions on continuous variables.
Remembering
file directories
In the File menu, the Add library option defaults
to the Analytica libraries directory. The Open Module and
Add Module options default to the current model directory,
usually where you last opened or saved the model. The data functions,
ReadTextFile and WriteTextFile, and menu options Export
and Import use the current data directory. You can change
these defaults with the File Finder dialog or the CurrentModelDirectory
or CurrentDataDirectory functions.
User
Guides
The Analytica User Guide 3.1 explains all features in Analytica
3.1, incorporating the previous Upgrade Guide 3.0. You can
access all documents electronically directly from the Help
menu as PDF using Adobe® Reader®.
What's
new in Analytica 3.1?