home
***
CD-ROM
|
disk
|
FTP
|
other
***
search
/
Simtel MSDOS 1992 September
/
Simtel20_Sept92.cdr
/
msdos
/
statstcs
/
mystat.arc
/
MYSTAT.HLP
< prev
next >
Wrap
Text File
|
1987-02-15
|
15KB
|
420 lines
@ 0
MYSTAT commands - for more information type HELP <command>
HELP information about commands EDIT create, edit MYSTAT file
INTRO print/display documentation and compute transformations
DEMO demo of MYSTAT in action SAVE save data in MYSTAT file
NAMES display variable names USE use data in MYSTAT file
PUT write data to ASCII file
PLOT two-way scatterplots WEIGHT weight cases by a variable
BOX boxplots SORT sort MYSTAT file
STEM stem-and-leaf diagrams RANK converts variables to ranks
TPLOT series plots
HISTOGRAM histograms SUBMIT submit commands from file
STATS descriptive statistics
TABULATE freqs, crosstabs, chi-square CHAR IBM or generic graphics chars
PEARSON correlation coefficients OUTPUT set output destination
TTEST independent/dependent t-tests NOTE print comment on output
SIGN sign test
WILCOXON Wilcoxon signed-ranks test FORMAT set # decimals to print
FRIEDMAN Friedman nonparametric ANOVA MENU use menu or just commands
KS 1/2 sample Komolgorov-Smirnov
CATEGORY define factors for ANOVA QUIT quit MYSTAT, return to DOS
MODEL specify regress/ANOVA model
ESTIMATE estimate model SYSTAT information about SYSTAT
@ 1
The PLOT command creates a two-way plot of one or more Y
variables on a vertical scale against an X variable on a
horizontal scale.
PLOT <var1>[,<var2>,<...>]*<var3> [/SYMBOL=<var$>|'<chr>'
YMAX=<#>,YMIN=<#>,XMAX=<#>,XMIN=<#>,LINES=<#>]
Examples:
PLOT Y*X
PLOT A*B / SYMBOL='*'
PLOT AGE*WEIGHT / SYMBOL=SEX$ (SEX$ case labeling variable)
PLOT Y1,Y2*X / SYMBOL='1','2' (Y1 values appear as '1', Y2 as '2')
PLOT Y*X / YMIN=0,YMAX=1,XMIN=1.5,XMAX=7.5
PLOT Y*X / LINES=100 (vertical size is 100 lines)
@ 2
The HISTOGRAM command creates a histogram for one or more
variables.
HISTOGRAM [<var1>,<var2>,<...> / BARS=<#>,SCALE,MIN=<#>,MAX=<#>]
Examples:
HISTOGRAM (histogram of every numeric variable in file)
HISTOGRAM X (histogram of X)
HISTOGRAM TRIAL(1),TRIAL(2)
HISTOGRAM TRIAL(1-5)
HISTOGRAM A,B / BARS=18 (forces 18 bars)
HISTOGRAM / SCALE (makes round cutpoints between bars)
HISTOGRAM / MIN=0,MAX=20 (set horizontal scale limits)
@ 3
The BOX command creates a boxplot for one or more variables.
BOX [<var1>,<var2>,<...>*<var3> / GROUPS=<#>,MIN=<#>,MAX=<#>]
Examples:
BOX (boxplot of every numeric variable in file)
BOX SALARY (boxplot of variable SALARY)
BOX TRIAL(1),TRIAL(2)
BOX TRIAL(1-5)
BOX SALARY*RANK (grouped boxplots of SALARY against RANK)
BOX INCOME*STATE / GROUPS=10 (plots of first 10 states only)
@ 4
The STEM command creates a stem-and-leaf diagram for one
or more variables.
STEM [<var1>,<var2>,<...> / LINES=<#>]
Examples:
STEM (stem-and-leaf of all numeric variables)
STEM TAX (stem-and-leaf of TAX)
STEM TRIAL(1),TRIAL(2)
STEM TRIAL(1-5)
STEM TAX / LINES=20 (20 lines used for diagram)
@ 6
The USE command specifies the MYSTAT file to use.
USE <file>
Examples:
USE MYDATA (reads from MYDATA.SYS in default drive/directory)
USE B:MYDATA (reads MYDATA.SYS from B: drive)
USE '\SYSTAT\NEWDATA.SYS'(fully qualified names must have .SYS extension)
@ 7
The SAVE command saves your data into a MYSTAT file. You must use
SAVE in EDIT or before RANK, SORT or ESTIMATE to create a file.
SAVE <file> [/ SINGLE | DOUBLE "<comment>"]
Examples:
SAVE FILE1 (saves data into FILE1.SYS)
SAVE FILE1 / SINGLE (saves data in single precision)
@ 9
The PUT command works like the SAVE command, except data are put
into a raw (ASCII) data file, not saved into a MYSTAT file.
PUT <file>
Example:
PUT RAWDATA (data will be put into RAWDATA.DAT)
@ 10
The TTEST command does dependent or independent t-tests.
TTEST [<var1>,<var2>,<...> [* <var3>]]
Examples:
TTEST A,B (dependent or paired)
TTEST A,B,C (dependent, each pair of variables)
TTEST A*GROUP (independent or two-group, GROUP has only 2 values)
TTEST A,B,C*GROUP (independent, three variables, two-group)
@ 11
The STATS command prints descriptive statistics. If you
choose no options, it produces N, MINIMUM, MAXIMUM, MEAN
and SD. Otherwise, it prints just the option(s) you choose.
Use BY to get subgroup statistics after sorting the file by
the grouping variable(s). BY must follow any statistics options.
STATS [<var1>,<var2>,<...>]
[/MEAN,SD,SKEWNESS,KURTOSIS,MINIMUM,MAXIMUM,RANGE,VARIANCE,SEM,SUM]
[BY <var3>,<var4>,<...>]
Examples:
STATS (basic statistics for whole file)
STATS VAR1,VAR2 / SEM (standard error of the mean)
STATS / BY GROUPS (basic statistics for cases in each group)
@ 12
The TABULATE command specifies an n-way table. With two-way tables
you get chi-square values and other statistics.
TABULATE <var1>[*<var2>*<...>]
[/FREQUENCY,PERCENT,ROWPCT,COLPCT,MISS]
Examples:
TABULATE AGE
TABULATE AGE*SEX (two-way table)
TABULATE AGE*SEX*STATE$ (three-way table)
TABULATE A,SEX*STATE$ (2 two-way tables)
TABULATE AGE*SEX / PERCENT (use percents in cells of table)
TABULATE AGE / MISS (ignore missing data)
@ 13
The SORT command sorts a file in ascending order on selected variables.
SORT <var1>,<var2>,<...>
Examples:
SORT RAINFALL (sort records in file by value of RAINFALL)
SORT NAME$ (alphabetic sort on variable NAME$)
SORT RAINFALL,STATE$ (nested sort)
First specify an input file with the USE command, then
an output file with SAVE, and finally the sorting keys
with SORT.
Example:
USE CITIES (input file)
SAVE SORTED (output file)
SORT RAINFALL (sort records in CITIES by value of RAINFALL)
@ 14
The RANK command converts variables to their ranks.
RANK [<var1>,<var2>,<...>]
Examples:
RANK (ranks all numeric variables in file)
RANK CLASS, INCOME
First specify an input file with the USE command, then
an output file with SAVE and finally the variables to
convert on RANK.
Example:
USE CITIES (input file)
SAVE SORTED (output file)
RANK (ranks all numeric variables in file)
@ 15
The WEIGHT command weights cases. The weighting variable
is reduced to an integer so that the sample size used in
the analysis is the sum of the weights. Cases weighted
less than one are excluded from the analysis.
An example is:
WEIGHT = COUNT
@ 16
The PEARSON command produces a matrix of Pearson product
moment correlation coefficients.
PEARSON [<var1>,<var2>,<...> / PAIRWISE | LISTWISE]
Examples:
PEARSON (matrix on all numeric variables in file)
PEARSON HEIGHT,WEIGHT
PEARSON TRIAL(1-5)
PEARSON / PAIRWISE (pairwise deletion of missing data)
PEARSON / LISTWISE (listwise deletion of missing data)
@ 17
The SUBMIT command sends a command file to MYSTAT for
processing as if you had typed the file from the console.
This file must be in your directory of files, must contain
characters (not binary data), and must have a CMD suffix.
You cannot nest SUBMIT files.
SUBMIT <file>
Examples:
SUBMIT COMMANDS (reads from file COMMANDS.CMD)
SUBMIT NEWJOB (reads from file NEWJOB.CMD)
@ 18
The OUTPUT command routes output to the console, a file,
or the printer.
OUTPUT * (sends subsequent output to console)
@ (sends output to printer)
<file> (sends output to a file)
Example:
OUTPUT FILE1 (sends output to FILE1.DAT)
@ 19
The NOTE command writes a comment on your output.
Example:
NOTE 'THIS IS A COMMENT.'
@ 21
The FORMAT command determines the number of digits to the
right of the decimal point in all numeric output. This
number cannot be larger than 9 nor less than 0. If you do
not use FORMAT, the standard value is 3.
Examples:
FORMAT=5
FORMAT=0
FORMAT=9/UNDERFLOW (print tiny numbers with exponents)
@ 22
HELP provides information about MYSTAT commands.
HELP [<command>]
Examples:
HELP (brief description of MYSTAT commands)
HELP STATS (description of STATISTICS command)
@ 24
The SIGN command computes a sign test on pairs of specified
variables.
SIGN [<var1> <var2> <...>]
Examples:
SIGN
SIGN TRIAL(1-2)
@ 25
The WILCOXON command computes a Wilcoxon signed-rank test
on pairs of specified variables.
WILCOXON [<var1> <var2> <...>]
Examples:
WILCOXON
WILCOXON TRIAL(1-2)
@ 26
The FRIEDMAN command computes a Friedman two-way analysis
of variance on selected variables.
FRIEDMAN [<var1> <var2> <...>]
Examples:
FRIEDMAN
FRIEDMAN TRIAL(1-10)
@ 27
The KS command computes one or two sample Kolmogorov-Smirnov
tests on specified variables.
KS [<var1> <var2> <...>] [/UNIFORM | NORMAL | CHISQ = <#>]
Examples:
KS (two sample tests on all possible pairs of variables)
KS TRIAL(1-2) (two sample tests on TRIAL(1) vs. TRIAL(2)
KS TOSS / UNIFORM (one sample test on uniform distribution)
KS Z / NORMAL (one sample standard normal)
KS DISTANCE / CHISQ=2 (one sample test on 2 df chi-square)
@ 28
The MENU command turns the MYSTAT menu on or off.
Examples:
MENU=ON
MENU=OFF
@ 31
The EDIT command starts the MYSTAT full screen editor.
EDIT [<filename>]
Examples:
EDIT (edit a new file)
EDIT CITIES (edit CITIES.SYS)
For further information, type EDIT [ENTER], [ESC], and then
type HELP.
@ 32
The CATEGORY command specifies the number of categories
for one or more variables used as categorical predictors (factors)
in an ANOVA or ANCOVA design. A CATEGORY variable must be numeric,
with values from 1 to k, where k is the number of categories specified.
CATEGORY <var1>=<#> [,<var2>=<#>,<...>]
Examples:
CATEGORY SEX=2 (SEX has 2 categories, with values 1 and 2)
CATEGORY GROUP=2,DRUG=3
@ 33
The MODEL command specifies a model to estimate.
MODEL <var1> [,<var2>,<...>] = [CONSTANT] + <var3>
[+ <...> + <var4>*<var5>]
Examples:
MODEL Y = X (no constant in model)
MODEL Y = CONSTANT + X (simple linear regression)
MODEL Y = CONSTANT + X + Z (multiple linear regression)
MODEL Y = CONSTANT + X + X*X + X*X*X (polynomial regression)
MODEL Y = CONSTANT + A + B + A*B (interaction term)
MODEL Y = CONSTANT + A + B + C + A*B + A*C + A*B*C (anova)
@ 34
The ESTIMATE command causes MYSTAT to estimate the previously
specified model.
@ 35
QUIT returns you to the operating system.
@ 36
@ 38
The TPLOT command plots a series of data values.
TPLOT [<var> / LAG=<#>,STANDARDIZE,MIN=<#>,MAX=<#>]
Examples:
TPLOT (plot first numeric variable)
TPLOT PRICE / LAG=10 (plot only first 10 cases of PRICE)
TPLOT VALUE / STANDARDIZE (standardize series in plot)
TPLOT EKG / MIN=0.,MAX=15. (set limits on plot scale)
@ 39
The INTRO command shows you how to use MYSTAT.
@ 40
SYSTAT The System for Statistics
Rated above SPSS/PC+ and PC-SAS in InfoWorld 9/1/86
Features in SYSTAT not found in MYSTAT:
SYSTAT BASIC programming language Principal components with rotations
RECODE statement for multiple codes and scores
File handling - match, merge, append Multidimensional scaling
Flexible input of ASCII files Cluster analysis, linkage or k-means,
Subgroup processing with SELECT & BY variables and/or cases
Loglinear models for contingency Differencing, detrending and
tables transformations of series
Repeated measures analysis Running means and medians
Stepwise regression ACF, PACF, Cross-correlation function
Extended regression diagnostics Seasonal and nonseasonal ARIMA
Canonical correlation Fourier analysis
Multivariate general linear model Nonlinear regression and estimation
Discriminant Analysis & Classification Logit, Probit, Rasch Item Analysis
Multi-way MANOVA, MANCOVA Telephone support and newsletter
SYSTAT runs on PC/MS-DOS, Apple Macintosh, VAX, CP/M, UNIX and other machines.
For more information about SYSTAT, please call (312) 864-5670 or write to
SYSTAT, Inc. 2902 Central Street Evanston, IL 60201
(SPSS/PC+ and SAS are registered trademarks of SPSS, Inc. & SAS Institute,Inc.)
@ 41
Type DEMO for an example of MYSTAT in action. This demonstrates the
full screen data editor and obtaining simple plots and statistics. It
also creates a file, CITIES.SYS, you can use to experiment with MYSTAT.
@ 42
The CHAR command allows you to select IBM screen/printer graphics
characters or generic characters that will print on any printer.
CHAR GRAPHICS | GENERIC
Examples:
CHAR GRAPHICS
CHAR GENERIC (use to plot on printers without IBM character graphics set)
@ 43
The NAMES command prints the names of the variables available to use.
Example:
NAMES
@