|
|
|
 |
|
|
計量経済分析用ソフトウェア
SHAZAM |
|
SHAZAM VERSION 8.0 の機能
SHAZAM is a comprehensive computer program for econometricians, statisticians, biometricians, sociometricians, psychometricians, politicometricians and others who use statistical techniques. Currently, SHAZAM is used in 77 countries from the Northernmost (University of Tromso, Norway) to the Southernmost (University of Otago, New Zealand) Universities in the world and in Antarctica. The primary strength of SHAZAM is for the estimation and testing of many types of regression models. The SHAZAM command language has great flexibility and provides capabilities for programming procedures. SHAZAM has an interface to the GNUPLOT package for high quality graphics.
SHAZAM の新しい機能：
* 時間変化する係数を有する回帰分析が行える
* ノンパラメトリク法の機能のサポート開始
* 一般化されたエントロピー法
その他の機能：
* data transformations, handling missing observations, matrix manipulation, evaluation of derivatives and integrals, data sorting, calculation of price indexes, computation of cumulative distribution functions for a variety of probability distributions;
* descriptive statistics, Dickey-Fuller and Phillips-Perron unit root tests, tests for cointegration;
* OLS estimation, restricted least squares, weighted least squares, ridge regression, distributed lag models, generalized least squares, estimation with autoregressive or moving average errors, estimation with heteroskedastic errors, ARCH and GARCH models, ARIMA (Box-Jenkins) time series models, Box-Cox regressions, probit models, logit models, tobit models, estimation using regression quantiles (including MAD estimation), regression with non-normal errors (including exponential regression and beta regression);
* linear and nonlinear hypothesis testing, calculation of confidence intervals and ellipse plots, computation of the Newey-West autocorrelation consistent covariance matrix, regression diagnostic tests (including tests for heteroskedasticity, CUSUM tests, RESET specification error tests), computation of p-values for many test statistics (including the p-value for the Durbin-Watson test), forecasting;
* nonlinear least squares, estimation of systems of linear and nonlinear equations by SURE, 2SLS and 3SLS, generalized methods of moments (GMM) estimation, pooled time-series cross-sectional methods;
* principal components and factor analysis, principal components regression, linear programming, minimizing and maximizing nonlinear functions, solving linear simultaneous equations.
Programming examples in the SHAZAM User's Reference Manual include splicing index number series, computing the power of a test, iterative Cochrane-Orcutt estimation, nonlinear least squares by the Rank One Correction method, Monte Carlo experiments, bootstrapping regression coefficients, Heteroskedastic Consistent Covariance Matrices, Hausman Specification test, non-nested model testing, solving nonlinear sets of equations, full information maximum likelihood, and multinomial logit models.
As well, examples of SHAZAM Procedures include square root of a matrix, Black-Scholes option pricing model, reproducing the OLS Command, limited information maximum likelihood, and generating multivariate random numbers.
|
|
| ←SHAZAMのTopページに戻る |
|
|
