|
|
 |
Setting the Standard¡¦
Until recently, using econometric software involved compromise. On
one hand, you could select software that provided you with powerful
statistical, forecasting, and modeling tools, but in a complex and
difficult-to-use package. Or you could choose software that featured
a modern, graphical interface, but was lacking features. It was your
choice: power and flexibility, or a user-friendly interface; state-of-the-art
modeling tools, or intuitive, easy-to-use software.
But that was then. With its innovative graphical object-oriented user
interface and sophisticated analysis engine, EViews has the power,
flexibility, and ease-of-use that you've been looking for. The only
choice for those who demand the very best, EViews is the worldwide
leader in Windows-based econometric software. Why compromise when
you can choose EViews? |
 |
A New Kind of User Interface
The goal in designing EViews was to make it both powerful and intuitive.
A wide range of statistical and graphical techniques had to be made
available without requiring users to memorize complicated command
syntax or navigate layers and layers of menus. The solution is an
innovative object-oriented user interface.
EViews is built around the concept of objects. Series, equations,
and systems are just a few examples of objects. Each object has its
own window, its own menus, its own procedures, and its own views of
its data. Most statistical procedures are simply alternative views
of the object. For example, a simple menu choice from a series window
changes the display between a spreadsheet, line and bar graphs, a
histogram-and-statistics view, a correlogram, and a unit root test.
Similarly, an equation window allows you to switch between a display
of the equation specification, basic estimation results, the coefficient
covariance matrix, graphics depicting the actual, fitted, and residual
values for the dependent variable, tables, forecast graphs and evaluations,
and more than a dozen diagnostic and hypothesis tests.
Naturally, you can cut-and-paste any of these views into your favorite
word processor with a simple menu selection. And it's just as easy
to exchange data and results with your spreadsheet and database programs.
EViews directly reads and writes an extensive list of data formats,
including Excel, ASCII/Text, SAS, Stata, SPSS, RATS, Html, Access,
Binary, ODBC Databases, ODBC queries (ODBC requires the Enterprise
version), and many others. To open most data files you can simply
drag-and-drop them onto EViews.
 |
|
Econometric Tools
Unlike some other econometric software, there is no reason for most
users to learn a complicated command language. EViews' built-in procedures
are a mouse-click away and provide the tools most frequently used
in practical econometric and forecasting work.
Basic Statistics
Basic descriptive statistics are easily computed over an entire sample,
by a categorization based on one or more variables, or by cross-section
or period in panel or pooled data. Hypothesis tests on mean, median
and variance may be carried out, including testing against specific
values, testing for equality between series, or testing for equality
within a single series when classified by other variables (which allows
you to perform one-way ANOVA).
You can graphically view the distribution of your data using histograms,
or cumulative distribution, survivor, and quantile plots. QQ-plots
(quantile-quantile plots) can be used to compare the distribution
of a pair of series, or the distribution of a single series against
a variety of theoretical distributions. You can even perform Kolmogorov-Smirnov,
Liliefors, Cramer von Mises, and Anderson-Darling tests to see whether
your series is distributed normally, or whether it comes from another
distribution such as an exponential, extreme value, logistic, chi-square,
Weibull, or gamma distribution. You may provide parameters for the
distribution, or let EViews estimate the parameters for you. EViews
also calculates kernel density estimates, and produces scatter plots
with curve fitting using ordinary, transformation, kernel, and nearest
neighbor regression.
To explore the time series properties of your data, EViews provides
unit root tests (ADF, Phillips-Perron, KPSS, DFGLS, ERS and Ng-Perron
for single time series and Levin-Lin-Chu, Breitung, Im-Pesaran-Shin,
Fisher, and Hadri for panel data), cointegration tests (with MacKinnon-Haug-Michelis
critical values and p-values), causality tests, autocorrelation and
partial autocorrelation functions, Q-statistics, and cross-correlation
functions.
EViews provides random number generators (Knuth, L'Ecuyer or Mersenne-Twister),
density functions and cumulative distribution functions for eighteen
different distributions. These may be used in generating new series,
or in calculating scalar and matrix expressions.
Seasonal Adjustment
EViews 5 provides easy-to-use front-end support for the U.S. Census
Bureau's X11 and X12-ARIMA seasonal adjustment programs, as well as
the Tramo/Seats software frequently used in Europe. Simple seasonal
adjustment using additive and multiplicative difference methods is
also supported in EViews.
Filters
EViews computes trends from time series data using the Hodrick-Prescott
filter. New to EViews 5 is the ability to apply Baxter-King, Christiano-Fitzgerald
fixed length and Christiano-Fitzgerald asymmetric full sample band-pass
(frequency) filters to your data.
|
|
Estimation
EViews includes a wide range of single and multiple equation estimation
techniques for both time series and cross section data. Basic estimators
include ordinary least squares (multiple regression), two-stage least
squares, and nonlinear least squares. Weighted estimation is available
with all of these techniques. Specifications may include polynomial
lag structures on any number of independent variables.
In addition to these basic estimators, EViews supports estimation
and diagnostics for a variety of advanced models.
EViews' sophisticated calculus engine computes and displays analytic
derivatives for the majority of nonlinear regression specifications.
ARCH models
If the variance of your series fluctuates over time, EViews can estimate
the path of the variance using a wide variety of Autoregressive Conditional
Heteroskedasticity (ARCH) models. EViews handles GARCH(p,q), EGARCH(p,q),
TARCH(p,q), PARCH(p,q), and Component GARCH specifications and provides
maximum likelihood estimation for errors following a normal, Student's
t or Generalized Error Distribution. The mean equation of ARCH models
may include ARCH and ARMA terms, and both the mean and variance equations
allow for exogenous variables.
Generalized Method of Moments
EViews supports GMM estimation for both cross-section and time series
data (single and multiple equation). Weighting options include the
White covariance matrix for cross-section data and a variety of HAC
covariance matrices for time series data. The HAC options include
prewhitening, either quadratic or Bartlett kernels, and fixed, Andrews,
or Newey-West bandwith selection methods.
Limited Dependent Variables
When your dependent variable takes on a limited set of values or is
censored or truncated, EViews can take account of this information
in the estimation procedure. Binary, ordered, censored, and truncated
models may be estimated for likelihood functions based on normal,
logistic, and extreme value errors. Count models may use Poisson,
negative binomial, and quasi-maximum likelihood (QML) specifications.
EViews optionally reports generalized linear model or QML standard
errors.
System Estimation
EViews supports estimation of both linear and nonlinear systems of
equations by OLS, two-stage least squares, seemingly unrelated regression,
three-stage least squares, GMM, and FIML. The system may contain cross
equation restrictions and autoregressive errors of any order.
Vector Autoregression/Error Correction Models
Vector Autoregression and Vector Error Correction models can be easily
estimated by EViews. Once estimated, you may examine the impulse response
functions and variance decompositions for the VAR or VEC. VAR impulse
response functions and decompositions feature standard errors calculated
either analytically or by Monte Carlo methods (analytic not available
for decompositions) and may be displayed in a variety of graphical
and tabular formats.
You may impose and test linear restrictions on the cointegrating relations
and/or adjustment coefficients. EViews' VARs also allow you to estimate
structural factorizations (VARs) by imposing short-run (Sims 1986)
or long-run (Blanchard and Quah 1989) restrictions. Over-identifying
restrictions may be tested using the LR statistic reported by EViews.
VARs support a variety of views to allow you to examine the structure
of your estimated specification. With a few clicks of the mouse, you
can display the inverse roots of the characteristic AR polynomial,
perform Granger causality and joint lag exclusion tests, evaluate
various lag length criteria, view correlograms and autocorrelations,
or perform various multivariate residual based diagnostics.
Panel Data Analysis and Pooled Time Series-Cross
Section
EViews features a wide variety of tools designed to facilitate working
with panel or pooled/time series-cross section data. Unbalanced or
balanced data sets with unlimited length time and/or cross-sections
are easily analyzed. In addition to ordinary linear and non-linear
least-squares, equation estimation methods include 2SLS/IV and Generalized
2SLS/IV, which can be used to estimate complex dynamic panel data
estimation including Anderson-Hsiao and Arellano-Bond types of estimators.
All of these methods allow both time and cross-section fixed and random
effect specifications. For random effects models, quadratic unbiased
estimators of component variances include Swamy-Arora, Wallace-Hussain
and Wansbeek-Kapteyn.
Also supported are AR specifications, weighted least squares, and
seemingly unrelated regression. Coefficients on specific variables
(including AR terms) can be constrained to be identical, or allowed
to differ across the cross-section.
State-Space Models
The state-space object allows estimation of a wide variety of single-
and multi-equation dynamic time-series models using the Kalman Filter
algorithm. Among other things, you can use the state-space object
to estimate random and time-varying coefficient models and ML ARMA
specifications.
Sophisticated procs and views give you access to powerful filtering
and smoothing tools so that you can view or generate one-step ahead,
filtered, or smoothed signals, states, or errors. EViews' built-in
forecasting procedures also provide easy-to-use tools for in- and
out-of-sample forecasting using n-step ahead or smoothed values.
User-Defined Maximum Likelihood Estimation
EViews 5 features an object (the LogL) for handling user-specified
maximum likelihood estimation problems. Simply use standard EViews
expressions to describe the log likelihood contribution of each observation
in your sample, and EViews will do the rest.
|
|
Specification Evaluation and Diagnostics
Once an equation or system is estimated, you can use EViews to perform
a wide array of specification evaluation and diagnostic tests.
These tests include Wald tests of linear and nonlinear coefficient
restrictions, likelihood ratio and F-tests for omitted variables,
Lagrange multiplier tests for serial correlation and ARCH, White heteroskedasticity
tests, Ramsey RESET tests, and Chow forecast and breakpoint tests.
Additional tests exist for specific models. As with other object views,
all hypothesis tests can be generated by a simple menu selection from
an equation or system window. |
|
Forecasting and Simulation
In EViews, you need not concern yourself with the complexities of
making forecasts. You can concentrate on the substance of the forecasting
problem. For single equation models, just select a menu item and EViews
will compute a static or dynamic forecast with optional forecast standard
errors and a graph of the 95 percent forecast confidence. Successful
forecasting equations can be saved in your workfile or stored in an
EViews database.
Simultaneous Equation Solution and Simulation
The model object, which is used for simultaneous equation simulation
and solution, provides the features most commonly requested by model
builders.
Variable dependencies and the block structure of the model¡¯s equations
are displayed with a simple mouse click. Reference equations by name
and the model is updated automatically whenever the equation is re-estimated.
You can even use the model to manage multiple solution scenarios for
comparing simulation results under various sets of assumptions.
The EViews model object makes it easy to perform non-stochastic or
stochastic simulation using either Gauss-Seidel or Newton solvers.
Built-in views and procedures display simulation results in graphical
or tabular form. Forward solution (currently unavailable with stochastic
solution) allows you to solve for model consistent expectations. EViews
provides sophisticated add factor support, including equation normalization.
You can even solve simple control problems where the values for an
exogenous control variable are found so that an endogenous variable
achieves a user specified target.
|
|
Data Management
Powerful modeling tools are only useful if you can easily access your
data. EViews provides the widest range of data management tools available
in any econometric software.
Extensive Function Library
EViews 5 contains an extensive library of functions for working with
and transforming your data. In addition to standard mathematical and
trigonometric functions, EViews provides functions for computing descriptive
statistics, by-group statistics, specialized date and time series
data functions, functions for working with a variety of statistical
distributions and date and string handling.
Sophisticated Expression Handling
EViews¡¯ powerful tools for expression handling mean that you can use
expressions virtually anywhere you would use a series. You don't have
to create new variables to work with the logarithm of Y, the moving
average of W, or the ratio of X to Y (or any other valid expression).
Instead, you can use the expression in computing descriptive statistics,
as part of an equation or model specification, or in constructing
graphs.
When you forecast using an equation with an expression for the dependent
variable, EViews will (if possible) allow you to forecast the underlying
dependent variable and will adjust the estimated confidence interval
accordingly. For example, if the dependent variable is specified as
LOG(G), you can elect to forecast either the log or the level of G,
and to compute the appropriate, possibly asymmetric, confidence interval.
Links, Formulas and Values Maps
Links allow you to create series that link to data contained in other
workfiles or workfile pages. Links allow you to combine data at different
frequencies, or match merge in data from a summary page into an individual
page such that the data is dynamically updated whenever the underlying
data change. Similarly, within a workfile, formulas can be assigned
to data series so that the data series are automatically recalculated
whenever the underlying data is modified.
Value labels (e.g., "High", "Med", "Low",
corresponding to 2, 1, 0) may be applied to numeric or alpha series
so that categorical data can be displayed with meaningful labels.
Built-in functions allow you to work with either the underlying or
the mapped values when performing calculations.
Data types
EViews can handle complex data structures, including irregular dated
data, cross-section data with observation identifiers, and dated and
undated panel data. In addition to numerical data, an EViews workfile
can also contain alphanumeric (character string) data and series containing
dates, which can be further manipulated using an extensive library
of functions.
In addition EViews provides a wide range of tools for manipulating
your data. Included is the ability to combine series by complex match
merge criteria. Workfile (dataset) procedures for changing the structure
of your data include: join, append, subset, resize, sort, and reshape
(stack and unstack).
File Import and Export
EViews provides extensive read/write support for foreign formats,
including Excel, ASCII/Text, SPSS, SAS (transport), Stata, Html, Microsoft
Access, Gauss Dataset, Rats, WinGive/PC Give, TSP, Aremos, dBase,
Lotus and Binary. Access to SAS native format files, version 8 or
earlier, is also available if a SAS ODBC driver is installed on the
system (which must be purchased separately through SAS).
EViews Databases
EViews 5 has built-in database features. An EViews database is a collection
of EViews objects maintained in a single file on disk. It need not
be loaded into memory in order to access an object inside it, and
the objects in the database are not restricted to being of a single
frequency or range. EViews databases support powerful query features
which can be used to search through the database for a particular
series or select a set of series with a common property.
Series contained in EViews databases may be accessed and used by EViews
procedures without being fetched into workfiles. Automatic search
capabilities allow you to specify a list of databases to be searched
when a series you need cannot be found in the current workfile.
Enterprise Edition Support for ODBC, FAME, DRIBase,
and Haver Analytics Databases
As part of the EViews Enterprise Edition (an extra cost option over
EViews Standard Edition), support is provided for access to data contained
in relational databases (via ODBC drivers) and to databases in a variety
of proprietary formats used by commercial data and database vendors.
Open Database Connectivity (ODBC) is a standard supported by many
relational database systems including Oracle, Microsoft SQL Server
and IBM DB2. EViews allows you to read or write entire tables from
ODBC databases, or to create a new workfile from the results of a
SQL query. For time series data, EViews also supports access to FAME
databases, both local and server based, Global Insight's DRIBase databases,
and Haver Analytics DLX databases. For time series databases, the
same, easy to use, EViews database interface is available no matter
what the source of the data.
Frequency Conversion
When you import data from a database or from another workfile, they
are automatically converted to the frequency of your current project.
EViews 5 has many options for frequency conversion, and includes support
for the conversion of daily, weekly, or irregular-frequency data.
Series may be assigned a preferred conversion method, allowing you
to use different methods for different series without having to specify
the conversion method every time a series is accessed. |
|
Graphics
EViews supports a wide range of graph types including line graphs,
bar graphs, filled area graphs, pie charts, scatter diagrams, mixed
line-bar graphs, high-low graphs, scatter plots and boxplots. A variety
of options give you control over line types, color, border characteristics,
headings, shading and scaling, including logarithmic scaling and dual
scale graphs. Legends are automatically created and you can add labels
in any scalable Windows font anywhere on your graph. Any number of
graphs can be combined in a single graph for presentation.
Customizing a graph is as simple as dragging graphic elements around
the screen. Want to change the characteristics of a legend or a text
label? Just click on it and your options are immediately presented
in easy to understand dialogs.
You can easily incorporate your customized graphs into other applications
using copy-and-paste or by writing the graph to a file. Formats supported
include Windows metafiles and PostScript files. |
|
Windows On-Line Help
Need help? EViews provides a full Windows-style help system with index
and search capabilities. In addition, the entire EViews User¡¯s Guide
and EViews Command and Programming Reference are provided in Adobe
PDF format (along with Adobe Acrobat Reader). Both manuals are extensively
hypertext linked, making it easy to find the information you need.
Heavily commented example programs (and sample data files) are indexed
to provide easy access to an array of expertly written EViews programs.
A Powerful Programming Language
Point-and-click is fine, but you feel more comfortable entering commands.
Besides, you need programming tools and capabilities. Well, EViews
is really two programs in one. In addition to its state-of-the-art
windowing interface, EViews includes a powerful command language that
allows access to all menu items.
Modelled loosely after the BASIC programming language but with new
object-oriented extensions and matrix handling capabilities. EViews
allows you to enter individual commands for immediate or batch execution.
Your programs can make use of advanced capabilities such as looping
and condition branching, as well as subroutine and macro processing.
Matrix primitives, from simple multiplication and inversion, to more
advanced procedures for Kronecker products, eigenvector solution,
and singular value decomposition, provide you with the tools you need
for solving your most complex problems. |
|
Data Capacity and System Requirements
EViews 5 supports most versions of the Windows Operating system including:
Windows 98/Me/NT 4.0/2000/XP. With sufficient memory in your computer,
you can tackle problems involving millions of observations or thousands
of series. The only fundamental capacity limit is that no single data
series or matrix may contain more than 4 million observations. And
because we take full advantage of 32-bit Windows¡¯ virtual memory,
you can work with data sets that exceed your system¡¯s physical memory. |
|
|
