9 Econometric software packages for financial and economic data analysis
None of modern financial econometrics projects can be executed without making use of an econometrics package. Following is a list of popular contemporary packages being widely, but not exclusively, used by financial economists and econometricians.
1 EViews
EViews, or Econometric Views, is a menu-driven and user-friendly package. The most recent version is EViews 6. It can easily handle most modern econometric models such as binary dependent variable models, univariate GARCH, cross section and panel data, and so on. Its help system in the electronic form is excellent, e.g., the Estimation Methods part provides detailed information on model specification and estimation, as well as the background and origin of the model.
However, the menu-driven feature also means that the package is not flexible to adapt to the need of specific requirements. Although there are many variations of GARCH available, they are all univariate.
The state space model and the Kalman filter can only do basic things, which are far from enough to cope with the requirement encountered in modern empirical studies featured by sophisticated model specifications and extensions. For more detail and purchase information visit the website given.
2. RATS
RATS, Regressional Analysis of Time Series, is one of the most authoritative packages in the area. With RATS version 7 there are many new features and improvements over the previous versions, such as more instrument variable/GMM support, and improved graphics.
Like EViews, RATS also has User’s Guide and Reference Manual in the electronic form, though it is an industry norm now. One of the advantages for using RATS is that, while a specialist package for time series analysis equipped with many readily executable procedures, the user can write or easily adapt a procedure for her/his own specific needs, or s/he can even write a procedure from scratch. Therefore, even if GARCH procedures were not provided, the user can write one with, e.g., RATS functions and the maximum likelihood procedure.
As such, virtually all kinds of contemporary time series models can be estimated with RATS, though sometimes it involves great complexity and requires much experience and skill. Mainly a time series package, one can also programme models of cross section and panel data with RATS. In addition to conventional analysis in the time domain, RATS can estimate time series in the frequency domain, also known as spectral analysis of time series.
Spectral analysis with RATS includes the Fourier transform, spectra and cross spectra, and coherence and phase. All of these are at the application level capable of handling empirical issues in business cycles and other problems involving cyclical movements and phase leads. The reader is recommended to visit Estima’s website where informative newsletters are published and useful procedures are logged and updated at some frequencies.
3 LIMDEP
LIMDEP, it is a complete econometrics package, as claimed by its developer and provider, Econometric Software, Inc. As such, LIMDEP Version 9.0 is an integrated programme for estimation and analysis of linear and nonlinear models, with cross section, time series and panel data. Since LIMDEP has long been a leader in the fields of discrete choice, censoring and truncation, panel data analysis, frontier and efficiency estimation, its collection of procedures for analysing these models is very comprehensive compared with other packages.
The main feature of the package is a suite of more than 100 built-in estimators for all forms of the linear regression model, and stochastic frontier, discrete choice and limited dependent variable models, including models for binary, censored, truncated, survival, count, discrete and continuous variables and a variety of sample selection models.
LIMDEP is widely used for teaching and research by higher education institutions, government departments and agencies, research establishments, business communities and industries around the world.
4 TSP
To some extent, TSP, or Time Series Processor, is similar to RATS. So we do not introduce it in detail and the reader can refer to the website of TSP International for detail.
5 GAUSS
GAUSS is powerful in matrix operations. The GAUSS Mathematical and Statistical System is a fast matrix programming language, one of the most popular software packages for economists and econometricians as well as for scientists, engineers, statisticians, biometricians, and financial analysts. Designed for computationally intensive tasks, the GAUSS system is ideally suited for the researcher who does not have the time required to develop programs in C or FORTRAN but finds that most statistical or mathematical ‘packages’ are not flexible or powerful enough to perform complicated analysis or to work on large problems. Compared with RATS, GAUSS is more power- ful and efficient but requires higher levels of programming knowledge and skills.
6. Microfit
http://www.econ.cam.ac.uk/microfit
Microfit is a menu-driven easy to use econometric package written especially for microcomputers, and is specifically designed for the econometric modelling of time series data. The strength of the package lies in the fact that it can be used at different levels of technical sophistication.
For experienced users of econometric programmes it offers a variety of univariate and multivariate estimation methods and provides a large number of diagnostic and non-nested tests not readily available in other packages. As a result, Microfit is one of the most frequently used econometric packages by economists and applied econometricians.
7. SAS
SAS is a large multi-purpose statistical package. It can process almost all model estimation problems. But as it is large it is not usually available on PCs. It also requires more knowledge in software.
8 Matlab
Matlab was initially developed for solving engineering problems. Now there are more and more economists and econometricians using this package. It is widely used for optimisation, control system design, signal and image processing and communications.
9 Mathematica
http://www.wolfram.com/products/mathematica
Economists and econometricians increasingly use this package as well. Its mathematics and algorithms cover matrices, algebra, optimisation, statistics, calculus, discrete mathematics and number theory.
With regard to statistics and data analysis, Mathematica provides integrated support both for classical statistics and for modern large-scale data analysis. Its symbolic character allows broader coverage, with symbolic manipulation of statistical distributions, symbolic specification of functions and models, and general symbolic representations of large-scale data properties.
Incorporating the latest numerics and computational geometry algorithms, Mathematica provides high-accuracy and high-reliability statistical results for datasets of almost unlimited size.
