RATS (statistical package)
Encyclopedia
RATS, an abbreviation of Regression Analysis of Time Series, is a statistical package for time series
analysis and econometrics
. RATS is developed and sold by Estima, Inc. (http://www.estima.com), located in Evanston, IL.
program called SPECTRE, written by Christopher Sims, a professor of economics. SPECTRE was designed to overcome some limitations of existing software that affected Sims' research in the 1970s, by providing spectral analysis and also the ability to run long unrestricted distributed lags. The program was then expanded by Tom Doan, then of the Federal Reserve Bank of Minneapolis
, who went on to found the consulting firm that owns and distributes RATS software. In its early incarnations, RATS was designed primarily for time series analysis, but as it evolved, it acquired other capabilities. With the advent of personal computers in 1984, RATS went from being a specialty mainframe
program to an econometrics
package sold to a much broader market.
RATS can read data from a variety of file formats and database sources, including Excel files, text files, Stata files, and most databases that support SQL and ODBC. It can handle virtually any data frequency, including daily, weekly, intra-day, and panel data.
RATS has extensive graphics capabilities. It can generate high-resolution time series graphs, high-resolution X-Y scatter plots, dual-scale graphs, and can export graphs to many formats, including PostScript and Windows Metafile.
New users often prefer the interactive mode, while experienced users will often prefer to run batch jobs. After an interactive session, the code can be saved, and converted to a batch format. One advantage of RATS, as opposed to automated forecasting software, is that it is an actual programming language, which enables the user to design custom models, and change specifications.
Recent versions have added report-generation tools designed to facilitate accurate exporting of results for use in papers and other documents.
. RATS has many of the same capabilities as SAS in both time series analysis and other advanced statistical methods. The two programs differ more in the details than in capabilities. SAS has routines for automated State Space estimation. RATS can be programmed to estimate State Space models, or regression models with time-varying coefficients. In this respect, RATS is actually more flexible. Similarly, SAS has an entire routine for estimating and forecasting with Unobserved Components Models. In RATS, estimation of this type would require extensive programming. Nevertheless, in general, the capabilities of RATS are comparable to SAS/ETS and SAS/STAT, but at a much lower price.
Time series
In statistics, signal processing, econometrics and mathematical finance, a time series is a sequence of data points, measured typically at successive times spaced at uniform time intervals. Examples of time series are the daily closing value of the Dow Jones index or the annual flow volume of the...
analysis and econometrics
Econometrics
Econometrics has been defined as "the application of mathematics and statistical methods to economic data" and described as the branch of economics "that aims to give empirical content to economic relations." More precisely, it is "the quantitative analysis of actual economic phenomena based on...
. RATS is developed and sold by Estima, Inc. (http://www.estima.com), located in Evanston, IL.
History
The forerunner of RATS was a FORTRANFortran
Fortran is a general-purpose, procedural, imperative programming language that is especially suited to numeric computation and scientific computing...
program called SPECTRE, written by Christopher Sims, a professor of economics. SPECTRE was designed to overcome some limitations of existing software that affected Sims' research in the 1970s, by providing spectral analysis and also the ability to run long unrestricted distributed lags. The program was then expanded by Tom Doan, then of the Federal Reserve Bank of Minneapolis
Federal Reserve Bank of Minneapolis
The Federal Reserve Bank of Minneapolis, located in Minneapolis, Minnesota, in the United States, covers the 9th District of the Federal Reserve, including Minnesota, Montana, North and South Dakota, northwestern Wisconsin, and the Upper Peninsula of Michigan...
, who went on to found the consulting firm that owns and distributes RATS software. In its early incarnations, RATS was designed primarily for time series analysis, but as it evolved, it acquired other capabilities. With the advent of personal computers in 1984, RATS went from being a specialty mainframe
Mainframe computer
Mainframes are powerful computers used primarily by corporate and governmental organizations for critical applications, bulk data processing such as census, industry and consumer statistics, enterprise resource planning, and financial transaction processing.The term originally referred to the...
program to an econometrics
Econometrics
Econometrics has been defined as "the application of mathematics and statistical methods to economic data" and described as the branch of economics "that aims to give empirical content to economic relations." More precisely, it is "the quantitative analysis of actual economic phenomena based on...
package sold to a much broader market.
Features
RATS is a powerful program, which can perform a range of econometric and statistical operations. The following is a list of the major procedures in econometrics and time series analysis that can be implemented in RATS. All these methods can be used in order to forecast, as well as to conduct data analysis. In addition, RATS can handle cross-sectional and panel data:- Linear regressionLinear regressionIn statistics, linear regression is an approach to modeling the relationship between a scalar variable y and one or more explanatory variables denoted X. The case of one explanatory variable is called simple regression...
, including stepwise. - Regressions with heteroscedasticity and serial-correlation correction.
- Non-linear least squaresNon-linear least squaresNon-linear least squares is the form of least squares analysis which is used to fit a set of m observations with a model that is non-linear in n unknown parameters . It is used in some forms of non-linear regression. The basis of the method is to approximate the model by a linear one and to...
. - Two-stage least squares, three-stage least squares, and seemingly unrelated regressions.
- Non-linear systems estimation.
- Generalized Method of Moments.
- Maximum likelihood estimation.
- Simultaneous equation systems, large econometric models.
- ARIMAArimaThe Royal Borough of Arima is the fourth largest town in Trinidad and Tobago. Located east of the capital, Port of Spain, Arima supports the only organised indigenous community in the country, the Santa Rosa Carib Community and is the seat of the Carib Queen...
(autoregressive, integrated moving average) and transfer function models. - Spectral analysis.
- Kalman filterKalman filterIn statistics, the Kalman filter is a mathematical method named after Rudolf E. Kálmán. Its purpose is to use measurements observed over time, containing noise and other inaccuracies, and produce values that tend to be closer to the true values of the measurements and their associated calculated...
and State Space models. - Neural networkNeural networkThe term neural network was traditionally used to refer to a network or circuit of biological neurons. The modern usage of the term often refers to artificial neural networks, which are composed of artificial neurons or nodes...
s. - Regressions with discrete dependent variables, such as logistic regressions.
- ARCHArchAn arch is a structure that spans a space and supports a load. Arches appeared as early as the 2nd millennium BC in Mesopotamian brick architecture and their systematic use started with the Ancient Romans who were the first to apply the technique to a wide range of structures.-Technical aspects:The...
and GARCH models. - Vector autoregressionVector autoregressionVector autoregression is a statistical model used to capture the linear interdependencies among multiple time series. VAR models generalize the univariate autoregression models. All the variables in a VAR are treated symmetrically; each variable has an equation explaining its evolution based on...
s.
RATS can read data from a variety of file formats and database sources, including Excel files, text files, Stata files, and most databases that support SQL and ODBC. It can handle virtually any data frequency, including daily, weekly, intra-day, and panel data.
RATS has extensive graphics capabilities. It can generate high-resolution time series graphs, high-resolution X-Y scatter plots, dual-scale graphs, and can export graphs to many formats, including PostScript and Windows Metafile.
Mode of Operation
RATS can be run interactively, or in batch mode. In the interactive mode, the user can run existing programs, or perform new tasks either by using menu-driven "wizards" or by typing in commands directly (or a combination of both approaches). The menu-driven wizards automatically generate the corresponding commands, allowing users to interactively construct complete programs that can be saved and re-run later.New users often prefer the interactive mode, while experienced users will often prefer to run batch jobs. After an interactive session, the code can be saved, and converted to a batch format. One advantage of RATS, as opposed to automated forecasting software, is that it is an actual programming language, which enables the user to design custom models, and change specifications.
Recent versions have added report-generation tools designed to facilitate accurate exporting of results for use in papers and other documents.
Comparison with Other Software
One advantage of the RATS program is that it is inexpensive, compared to larger programs such as SASSAS System
SAS is an integrated system of software products provided by SAS Institute Inc. that enables programmers to perform:* retrieval, management, and mining* report writing and graphics* statistical analysis...
. RATS has many of the same capabilities as SAS in both time series analysis and other advanced statistical methods. The two programs differ more in the details than in capabilities. SAS has routines for automated State Space estimation. RATS can be programmed to estimate State Space models, or regression models with time-varying coefficients. In this respect, RATS is actually more flexible. Similarly, SAS has an entire routine for estimating and forecasting with Unobserved Components Models. In RATS, estimation of this type would require extensive programming. Nevertheless, in general, the capabilities of RATS are comparable to SAS/ETS and SAS/STAT, but at a much lower price.