| GenStat for Windows Product Details
Brief Description of the GenStat System GenStat
is a GENeral STATistics package developed by VSN International Ltd.
GenStat is a comprehensive statistics system which
offers ease-of-use for the novice user through a Windows menu interface,
or power and flexibility for the more experienced user through a powerful command language interface.
One of the main strengths of GenStat is the range of powerful statistics techniques
available which have been tested time and again by practicing statisticians. The range of
statistics available includes basic statistics, design and analysis of designed
experiments, analysis of linear and generalized linear mixed models, microarray analysis, regression (linear, nonlinear and generalized linear), hierarchical
generalized linear models, spatial analysis, multivariate analysis techniques, time series, statistical process control methods,
survival analysis, sample size calculations and resampling methods.
Go to main list
GenStat for Windows With GenStat for Windows, simple and complex analyses alike are at your finger tips
with the extensive pull-down menu system. GenStat menus have a consistent ANALYSE - CHECK
ASSUMPTIONS - SAVE DETAILS approach throughout to help you think logically and interactively
about your data analysis. Should you wish to use the powerful GenStat command language instead of menus -
no problem - simply open a new text window, write the command you need and submit them to GenStat.
Tracking your session is easy with the Audit Trail; a
translation of your menu analysis into GenStat code. This can be used as proof of the
steps taken in your analysis, or run at a later date as a time-saving GenStat program.
With the SAVE SESSION facility you can take a snapshot of your GenStat session at any time and
restart it at a later date - no need to work back through the analysis. SAVE SESSION records
all the available data in GenStat at any given point, all open menus and their associated settings,
all open spread sheets and all open text windows/GenStat command programs.
GenStat for Windows 10th Edition
To achieve the best with the current version, the recommended system configuration required is an IBM compatible PC with:
- Pentium or compatible processor
- Recommended minimum 64 MB RAM
- Windows 98, ME, 2000 or XP
- 140 MB Hard disk space (maximum disk space required)
- 80 MB Hard disk space for Microarray examples
Go to main list
Why Choose GenStat? GenStat is a powerful and flexible statistical system developed by statisticians at the
Rothamsted Experimental Station, where many standard statistical techniques were first
developed.
The GenStat system offers you the benefits of a fully interactive system with a user
friendly graphical interface plus extensive graphical facilities. It
has a long history of successful use and is continuously updated and improved to ensure
that it remains at the fore-front of statistical technology.
Unlike many other systems that are sold in a modular fashion, GenStat is a complete package.
There are no add-on extras to pay for; GenStat's considerable range of analyses is available to all users.
The graphical interface and the logical, straightforward command language make
the power of GenStat readily available to the non-statistician and statistician
alike.
The GenStat system has something to offer everyone who uses statistics especially in
the following areas:
- scientific research
- financial statistics
- engineering
- social sciences
- statistical education
- medical statistics
- pharmaceutical experiments
- agricultural experiments
Go to main list
The GenStat Approach GenStat is an extremely powerful toolkit for all your statistical needs and is
designed to be customizable. With GenStat you can manipulate, visually explore or
analyse your data using a full range of techniques; from simple summary statistics and plots to complex
nonlinear modelling.
The flexibility of GenStat allows you to use a dynamic, interactive approach to
modelling, easily trying out different models and examining results. GenStat also allows
you the freedom to explore new models, adapt many of the models supplied, or even develop
your own.
In GenStat for Windows, data can be input directly from external spread sheets, databases and other
mathematical and statistical packages. A comprehensive range of statistical analyses can be
performed via the user friendly pull-down menu system.
Go to main list
Data Formats Supported
GenStat for Windows users can read data from the following formats directly into GenStat:
| ASCII |
Lotus |
Minitab |
| dBase |
Excel |
SAS |
| Quattro |
ODBC data & retrieval files |
SPSS |
| Paradox |
|
S+ |
| Matlab |
|
Systat |
| ArcView |
|
MSTAT |
| MapInfo |
|
Stata |
| Gauss |
|
INSTAT |
| Windows Bitmap |
|
Epi-Info |
| Wave Sound Files |
|
Statistica |
| |
|
Sigma Plot |
| |
|
OSIRIS |
| |
|
R |
(Non-windows GenStat users can use the IMPORT command with a greater range of data formats in a non-Windows environment.)
Go to main list
Statistical Facilities Basic Statistics
Exploratory Data Analysis
Good statistical practice means that the first steps of any analysis should be the familiarization of the experimental
data, to check that it has been carefully collected, or to graphically explore its nature. To achieve this, GenStat offers a
range of tools for describing and visualizing data. Summarize Contents of Variates Menu
The Summarize Contents of Variates Menu is often the first stop for exploring
your continuous data structures. A range of statistics can be produced: number
of values, number of missing values number of non-missing values, minimum,
maximum, range, median, quartiles, total of values, variance, standard deviation,
arithmetic mean, standard error of mean, coefficient of variation, corrected
sum of squares, uncorrected sum of squares, skewness, standard error of skewness,
kurtosis, s.e. of kurtosis.
Convenient exploratory graphs can also be produced here to visually explore your data:
- Histogram
- Box-and-whisker plots
- Normal plots
- Stem and leaf plots
NB: Many more graphical displays are available through the Graphics menu
Tally Menu
The Tally menu forms a simple tally table of the distinct values in a vector.
The tally table of a vector, gives the counts, percentages, and cumulative counts and percentages of each distinct value.
Frequency Tables Menu
This menu produces frequency counts for grouped data. The data for this menu is
given by one or more factors and the frequency counts are then the number of units
in each factor combination. The menu also includes facilities for producing tables
classified by multiple response factors.
GenStat's Copy Special facility allows you to easily save your tables to RTF and HTML formats
for easy inclusion in documents and web pages.
Summaries of Groups (Tabulation) Menu
This menu produces summaries of data values classified into different groups.
GenStat's Copy Special facility allows you to easily save your tables to RTF and HTML formats
for easy inclusion in documents and web pages.
Summarize Circular Data
This menu provides facilities for summarizing data values that consist of
directional observations recorded as angles between 0 and 360 degrees. Two tests
of uniformity are calculated. The null hypothesis for both of these is that the
observations come from a uniform distribution around the circle.
Correlations Menu
The Correlations menu allows you to easily produce and compare the correlations between
your continuous data structures.
Go to main list Statistical Tests
A range of statistical tests are available through GenStat menus:
- T-Test - one-sample & two-sample (paired and unpaired) t-tests
- Binomial test - one-sample & two-sample
- Poisson test - one-sample & two-sample
- One-sample non-parametric tests (Wilcoxon, Runs and Sign tests)
- Two-sample non-parametric tests ( Wilcoxon matched pairs test, Mann-Whitney U test, Kolmogorov-Smirnoff test,
Sign test, Spearman's Rank Correlation)
- Kendall's Coefficient of Concordance
- Kruskal-Wallis One-way ANOVA
- Friedman's Non-parametric ANOVA
- Kappa Statistic
- Gamma Statistic
- MANTEL test
- Contingency Tables (Chi-square test, Fisher's exact test for 2x2 table of counts)
- McNemar's test
- Lin's concordance coefficient
NB: More tests are available outwith the menu system.
Go to main list
Fitting Distributions
Probability distributions may be examined and parameters of a range of distributions
may be estimated through the Distributions menu, including:
| Discrete |
|
Continuous |
| Poisson-Pascal |
|
Double Normal (equal variances) |
| Poisson |
|
Double Normal (unequal variances) |
| Geometric |
|
Log-Normal |
| Log-series |
|
Exponential |
| Negative binomial |
|
Gamma |
| Neyman type A |
|
Weibull |
| Pólya-Aeppli |
|
Beta type I |
| Poisson-log-Normal |
|
Beta type II |
| |
|
Pareto |
Go to main list
The Design of Experiments Comprehensive facilities for designing experiments are
available. Collectively known as the GenStat Design System, these design facilities are
intended to provide a coordinated set of facilities for the selection and construction of
effective experimental designs, aiming to give:
- a non-technical interface for the general user (Windows) and a faster (command-based) method for
the experienced user.
- a good range of standard designs but without constraining users to this predefined set:
- One-way Design (no Blocking)
- One-way Design (in Randomized Blocks)
- Two-way Design (no Blocking)
- Two-way Design (in Randomized Blocks)
- General Treatment Structure (no Blocking)
- General Treatment Structure (in Randomized Blocks)
- Completely Randomized Design
- Split-Plot Design
- Latin Square
- Graeco-Latin Square
- Split-Split-Plot Design
- General Split-Plot Design
- Lattice Design
- Strip-plot (or Criss-cross) Design
- Factorial Design with Added Control
display facilities for designs, plans and data forms
manipulation procedures to combine simple designs
Go to main list
The Analysis of Designed Experiments
One of the pillars of statistical computation is Analysis of Variance, and
GenStat contains a unique and powerful algorithm which allows analysis of
all generally balanced designs. The simple specification of a design in the
GenStat menu system belies the enormous scope of designs that GenStat can handle.
Any balanced design can be specified through the General Analysis of Variance menu, whilst specific ANOVA menus
give you the extra bit of help you may need for:
- One-way ANOVA (no Blocking)
- One-way ANOVA (in Randomized Blocks)
- Two-way ANOVA (no Blocking)
- Two-way ANOVA (in Randomized Blocks)
- Completely Randomized Design
- Split-Plot Design
- Split-split Plot Design
- Latin Square
- Graeco-Latin Square
- Lattice Design
- General Treatment Structure (no Blocking)
- General Treatment Structure (in Randomized Blocks)
- Unbalanced Treatment Structure
In the case of unbalanced data, GenStat's flexible regression facilities may also be used for Analysis of Variance,
or the much more powerful, REML facilities.
Go to main list
The Analysis of Mixed Models and Unbalanced Designs
With the powerful REML algorithm, GenStat offers state-of-the-art facilities,
second to none, for the analysis of:
- Linear mixed models
- Repeated measurements
- Multivariate linear mixed models
- Random coefficient regression
- Spatial models (regular and irregular grids)
- Multiple Experiments/Meta Analysis
- Generalized linear mixed models
All available through the GenStat menu system or command language
Go to main list
Spatial Analysis Spatial Statistics embodies a suite of techniques for analysing data distributed in a space
of one, two or three dimensions and for estimating (predicting, kriging) local values in
that space. The Spatial Analysis menus in GenStat are:
- Form Variogram - forms auto variograms for individual variates or cross variograms for pairs
of variates
- Model Variogram - this menu uses the directives FIT, FITCURVE and FITNONLINEAR to fit
various models to the experimental variogram
- Krige - calculates kriged estimates using a model fitted to the sample variogram.
Go to main list
Microarray Analysis Microarray experiments are used in genetics to screen RNA samples against
libraries of DNA/RNA samples/fragments. The microarray menus provide facilities
for the design and analysis of microarray experiments:
- input microarray data
- calculate log ratios
- empirical Bayes estimates
- false discovery rates
- visualize
- normalize
- analyze
- display
- cluster
Go to main list
Regression Analysis
Within GenStat, regression means much more than fitting straight lines. Certainly, lines can be fitted very simply (Linear Regression Menu); but GenStat
is designed to make it easy to compare alternative lines, fitted either to different groups of the observation, or using different
explanatory variables. Lines do not have to be straight: you can fit polynomial curves, choose from a range of standard
nonlinear curves (Standard Curves Menu), or even define your own curves and models (Nonlinear Models Menu).
You can choose from a range of distributions for the response variable and this,
together with a range of link functions, allows you to analyse what are known
as generalized linear models. If you wish to expand GenStat's generalized linear
facilities even further, you can also define your own distributions and link
functions.
Also include:
- Hierarchical Generalized Linear models; these extend generalized linear models (GLMs) to include additional random terms in the linear predictor.
- Screening Tests
- Regression Trees
For each class of regression in GenStat, the following are available:
- regression with factored variables
- weighted and offset regression
- polynomial and non-parametric regression
- plots for model checking
- plots of fitted models
One of the strengths of GenStat is that a complex generalized linear model is as straightforward to fit to data as that of a
simple linear regression. Like the ANOVA menus, GenStat regression menus give you the opportunity to select from a 'general menu'
where any regression model can be specified, or specific menus where more help is given (e.g. multiple linear regression with groups,
probit analysis, logistic regression).
GenStat regression menus are designed to make model selection as interactive as possible; fit a model and then use the Change Model
to Add, Drop or Try additional explanatory variables - the choice is yours. For linear and generalized linear modelling you can
get extra help from GenStat through the Stepwise Regression menu.
Go to main list
Multivariate Analysis Many standard and some more specialised multivariate methods are available through the
GenStat menu system:
- Principal components
- Canonical variates analysis
- Principal coordinates
- Multidimensional scaling
- Hierarchical and non-hierarchical cluster analysis
- Procrustes rotation
- Correspondance analysis
- Canonical correlations
- Multivariate analysis of variance
NB: Many more MV techniques are available through the
GenStat command language - contact us for more details.
Go to main list
Survival Analysis The GenStat menu system offers a range of tools for Survival analysis:
- The Kaplan-Meier menu calculates the Kaplan-Meier estimate of the survivor
function, using data recorded at the end of specified time-intervals.
- The Survival Distributions menu models survival times.
- The Proportional Hazards menu fits the proportional hazards model to survival data.
Go to main list
Six Sigma
Six Sigma facilities embodies a suite of menus to help
keep your industrial processes working smoothly and efficiently:
- Control Charts for Measurements: This menu plots a range of standard
control charts for measurements. The type of charts available include the
standard charts devised by Shewhart (1931) for the control of manufacturing
processes, exponentially weighted moving-average control charts for
controlling the mean of a process and cumulative sum (or CUSUM) charts.
- Control Charts for Attributes: This menu plots a range of standard control
charts for attributes. The type of charts available include the c or u
charts representing numbers of defective items and p or np charts for
binomial testing for defective items.
- Pareto Charts: This menu sorts tables so that their margins are in a
specified order. With a multi-way table, for example, this may help in
interpreting an interaction from an analysis of variance. With a one-way
table, it allows the cells to be displayed in ascending order, as in a
Pareto chart.
- Capability Statistics: This menu calculates capability statistics.
These are used to assess the extent to which the output of a process lies
within its specification limits. The data values consist of samples of
measurements made on successive occasions.
- Industrial Designs: This menu helps you generate a range of designs
commonly used in industry:
- factorial designs (with interactions confounded with blocks)
- fractional factorial designs (with blocking)
- balanced-incomplete-blocks
- central composite designs
- Box-Behnken designs
- Plackett Burman (main effect) designs
Go to main list
Sample Size Calculations
- t-tests
- Binomial tests
- Sign tests
- McNemar's test
- Mann-Whitney tests
- Lin's concordance coefficient
- Correlation coefficients
Go to main list
Survey Analysis
The GenStat menus give you the tools to make sense of your survey data:
- Tally: The Tally menu forms a simple tally table of the distinct values in a vector. The tally table of a vector, gives the counts, percentages, and cumulative counts and percentages of each distinct value.
- Frequency Tables: This menu produces frequency counts for grouped data. The data for this menu is given by one or more factors and the frequency counts are then the number of units in each factor combination. The menu also includes facilities for producing tables classified by multiple response factors.
- Summaries of Groups (Tabulation): This menu produces summaries of data values classified into different groups.
- Design-based Survey Analysis: This menu is used to do a simple design-based analysis for a simple or stratified random survey.
- Ratio analysis for surveys: This menu is used to do a ratio analysis for a simple or stratified random survey.
Go to main list
Time Series Analysis Individual time series models can be modelled by BoxJenkins ARIMA or seasonal ARIMA
models and may be extended to general transfer functions between series.
The Time Series/ Data Exploration menu Displays time series statistics useful for
ARIMA model selection. It also produces numerical summaries and graphical output. The
Time Series/ARIMA Model Fitting menu fits an autoregressive integrated moving-average (ARIMA)
model to time-series data. The series and results can be displayed graphically and
forecasts of future observations can be formed.
A wide range of facilities for fitting more complex
Box-Jenkins models (including regression with correlated errors or transfer function
models) is available in command mode. This includes several procedures for graphical
exploratory analysis of time series data, including spectral analysis. Commands
also enable filtering of time series and calculation of Fourier transforms for use
in spectral analysis.
Go to main list
Matrix Operations
The matrix is an important structure in statistics, especially in the case of multivariate techniques. As a result, GenStat
supports matrices as a specific structure type and offers a considerable range of tools for matrix calculation and
manipulation:
- matrix arithmetic - addition, subtraction multiplication
- transpose
- invert
- determinant of a square matrix
- trace of a square matrix
- form the sums of squares and products matrix from several vectors
- correlation matrix derived from a symmetric matrix
- singular value decomposition
- eigenvalue decomposition
- Choleski decomposition of a matrix
- solution of simultaneous linear equations
The structure of GenStat for Windows is such that data calculations may be carried out through in the GenStat
spread sheet facilities or through the Calculate menu. GenStat's matrix facilities are no exception.
Go to main list
Graphics
GenStat offers state-of-the-art 2D and 3D graphics where you can identify
observations, zoom, spin and edit interactively.
Choose from short-cut menus or go via the Graphics Wizard to create:
- 2D Scatter Plots
- 2D Line Plot
- 2D Line & Scatter Plot
- 2D Histogram
- Barchart
- Boxplot
- Rug Plot
- Piechart
- Dot Plot
- Minimum Spanning Tree
- Contour Plot
- Surface Plot
- Shade Plot
- 3D Histogram
- 3D Scatter Plot
- Trellis Plot
- Repeated Measures Profile Plot
- Windrose Diagram
- Circular Plot
- Biplot
In addition, analysis specific diagnostic plots are produced from statistics menus where appropriate. For example:
- Plots of fitted lines in linear regression
- Residual, normal, half normal and histograms of residuals for ANOVA and regression
- Plot/treatment layout for a designed experiment
- Dendrograms for cluster analysis
- Time series plots
- Six sigma control charts
... and more.
Once you have created and edited GenStat graphics to your satisfaction, you can import your graphs into
Word Processor documents and presentations in the following formats:
-
JPEG
-
GIF
-
Portable Network Graphics (PNG)
-
PostScript
-
Encapsulated PostScript
-
HPGL
-
Windows enhanced metafiles
-
Bitmap
-
Computer graphics metafiles (CGM)
Go to main list
Extending GenStat
GenStat for Windows is not simply a tool for the novice or inexperienced user. GenStat
boasts a powerful command language that is a programming language in its own right. This
not only gives you the flexibility to work how you like, but also to extend the facilities
available through procedure writing.
A Procedure is a GenStat program is like a macro, i.e. it is a program written in GenStat
language that operates in the same way as a single GenStat command. Why not visit the
GenStat User Area on our web site
to see the kind of functionality other users have added to GenStat using this method?
Once you are familiar with the GenStat language, you can
-
Build and attach your own procedure libraries.
-
Add your own menus:
Component Object Model (COM) add-ins are supplemental programs that provide
additional functionality in a variety of programming languages, including
Visual Basic, Visual C++, and Visual J++. Within GenStat this can be used to
create dialogs to run a set of GenStat commands, thus building customized menus.
Go to main list
Availability GenStat is available for a wide range of systems including UNIX workstations.
Please contact us for further details.
Go to main list
Discussion List GenStat has its own electronic bulletin board, monitored by the GenStat
developers, where users can share experiences and ideas. To subscribe to the list,
send the message
SUBSCRIBE GenStat first-name last-name
to the address:
LISTSERV@LISTSERV.RL.AC.UK |
RSS Feeds