home
***
CD-ROM
|
disk
|
FTP
|
other
***
search
/
Simtel MSDOS 1992 September
/
Simtel20_Sept92.cdr
/
msdos
/
statstcs
/
pdswstat.arc
/
PDSWSTAT.LST
Wrap
File List
|
1989-09-25
|
71KB
|
1,556 lines
PUBLIC DOMAIN (FREEWARE) / USER SUPPORTED (SHAREWARE) STATISTICAL SOFTWARE
AND RELATED UTILITY PROGRAMS
Review by Jim Groeneveld, September 1989.
-----------------------------------------------------------------------------
NIPG TNO - - - - - <work> - - - - -|- <home> - - - - -| GROENEVELD@HDETNO51
Postbus 124 | Wassenaarseweg 56 | Schoolweg 14 | GROENEVELD@TNO.NL
2300 AC Leiden | 2333 AL Leiden | 8071 BC Nunspeet | TNOSUR::GROENEVELD
Nederland (NL) 071-178810 | 03412-60413 | RULTNO::JIM
-----------------------------------------------------------------------------
This is a describing inventory, NOT a qualitatively and quantitatively
comparative research. NOT included are several statistical applications in
dBASE or LOTUS. Neither claims this review to be complete.
Distribute this list without changes. Send comments, suggestions and
additions to one of the above addresses to maintain uniform updates.
SUMMARY
Except for some small programs
per software package the following data are included: (X = sequence number)
X. name, vs / date / author / system / pd-sw [price] / summarized description
A. Multi-purpose (& multi-program) statistical packages:
1. STAT / 11.26.81 / ats / IBMPC / pd / basic stats with limited data (<50?)
2. EPISTAT 3.0/ 1984 / Tracy L. Gustafson / IBMPC & BASICA / sw $25 / basic stat
3. SPPCdemo 1.0 / 1984 / WALMYR Publishing Co. / IBMPC / sw $75 / statistics
4. KWIKSTAT 1.0 / 1986 / Alan C. Elliott / IBMPC + CGA / sw $35 / statistics
5. SuperSTat (SST) 1.00 / 1986 / Mike Derby / IBMPC / sw $35/$20 / market stats
6. RA 1.5 / Jan. 1987 / Mark E. Williams / IBMPC / sw $45 / simple statistics
7. STATistiek / 1987 / Piet van Blokland / IBMPC / sw f50 / enquete & statistiek
8. TSxSTyy / 27 Nov.1988 / Timo Salmi / IBMPC&CGA /pd/ statistical programs
9. MAPSTAT 2.0 / 1988 / Douglas L. Anderton / IBMPC / pd / statistical package
B. Single-purpose, separate statistical programs:
1. GRAPH2 2.0 / May 20 1984 / Bob Shannon / IBMPC + CGA & BASICA / pd $20 / plot
2. SPLINE 2.01 / October 1986 / D.A.A.N. / IBMPC+CGA & GWBASIC / smooth curve
3. CURVEFIT 1.00 / May 21 1984 / Thomas S. Cox / IBMPC / pd / curve fitting
4. REGRESSI // Keith D. Barber / IBMPC / pd / linear & geom. & exp. regression
5. REGRESS / 4-11-83 / Engineering Software Inc. / IBMPC / pd / curve fitting
6. SORBUS 1.0 / 01-21-87 / F.J.H. Lupschen / IBMPC / pd / bar charts
7. STATPROG / 1986/87 / Joseph C. Hudson / IBMPC / pd $5/- / statistics progs.
8. STATOOLS / 1985/86 / Gerard E. Dallal / IBMPC / pd / statistics utilities
9. NCCLSPRE 1.83 / N. Adams / IBMPC&BASICA / pd / calculates a samples precision
10. NORMAL-Z / 3-3-82 / Craig W. Uthe / IBMPC&BASICA /pd/ normal distribution `Z'
11. HISTGRAM (displays and prints simple histogram in text mode) / IBMPC & BASICA
12. PIECHART (some extension of IBM's PIECHART) / IBMPC+CGA&BASICA /pd/ {English}
13. PIE2 (about the same as PIECHART) / IBMPC+CGA & BASICA / pd / {BASIC/English}
14. BAR's (several separate programs to generate bar charts) / IBMPC+CGA&BASICA /
15. BARCHART (simple bars, max. 4 per grouping, text mode, graphic characters) /
C. Function plotting programs (based on specified functions, not on data points):
1. GRAFIEK / 1988 / Piet van Blokland e.a. / IBMPC + CGA / sw f100 / plot func's
2. PLOTF(X) 0.1 / 1986 / Jim Groeneveld / IBMPC + CGA & BASICA / pd / func-plot
3. FPLOT /1983 / David Bush / IBMPC+CGA & GWBASIC / pd / Cartesian/Polar/3D-plot
4. PCGRAF /1985 / Henning Mittelbach / IBMPC+CGA & BASICA / pd / 3D-surface plot
5. 3D-PLOT /1986 / W. van Huissteden / IBMPC+Hercules&HBASIC / pd / surface plot
D. Statistical Pascal routines:
1. STAT-PAS / May 1985 / Philip Burns / IBMPC / pd / statistical Pascal routines
2. PASCALIB 1.00 / 1987 / N.D. Mackay / IBMPC / sw f25 / a.o. stat. routines
E. Specific data manipulation programs/utilities:
1. SEPARATE / 1987 / Jim Groeneveld / IBMPC / pd / data reformat/split
2. COPYFIX 1.0 / 1989 / Jim Groeneveld / IBMPC / pd / data (asciitext) reformat
3. CODEBOOK 1.0 / 1989 / Jim Groeneveld / IBMPC / pd / data reformat/reduction
PUBLIC DOMAIN (FREEWARE) / USER SUPPORTED (SHAREWARE) STATISTICAL SOFTWARE
AND RELATED UTILITY PROGRAMS
Review by Jim Groeneveld, September 1989.
-----------------------------------------------------------------------------
NIPG TNO - - - - - <work> - - - - -|- <home> - - - - -| GROENEVELD@HDETNO51
Postbus 124 | Wassenaarseweg 56 | Schoolweg 14 | GROENEVELD@TNO.NL
2300 AC Leiden | 2333 AL Leiden | 8071 BC Nunspeet | TNOSUR::GROENEVELD
Nederland (NL) 071-178810 | 03412-60413 | RULTNO::JIM
-----------------------------------------------------------------------------
This is a describing inventory, NOT a qualitatively and quantitatively
comparative research. NOT included are several statistical applications in
dBASE or LOTUS. Neither claims this review to be complete.
Distribute this list without changes. Send comments, suggestions and
additions to one of the above addresses to maintain uniform updates.
Except for some small programs
per software package the following data are included: (X = sequence number)
X. name, vs / date / author / system / pd-sw [price] / summarized description
a) name
b) version
c) date
d) programming language
e) user interface language
f) system requirements ('IBM' includes 'compatible' unless otherwise mentioned)
g) PD/SW disk number(s) / distribution point (BBS tel.nr., EARN/BITNET address)
h) public domain / shareware & price
i) author(s) / companies
j) address (for contribution) / telephone
k) description / purpose
l) summary of contents
m) additional/special features
A. Multi-purpose (& multi-program) statistical packages:
1. STAT / 11.26.81 / ats / IBMPC / pd / basic stats with limited data (<50?)
a) STAT
b) -
c) 11.26.81
d) BASICA / GWBASIC
e) English
f) IBMPC & BASICA / GWBASIC
g) PCSIG#180
h) pd
i) ats (whoever that may be)
j) unknown
k) set of 18 BASIC programs (Sx.BAS) chained by menu program (STAT.BAS);
data entry from keyboard (generally up to 50 values/pairs-?)
l) 1 . PERMUTATIONS AND COMBINATIONS
2 . MANN-WHITNEY U TEST
3 . GEOMETRIC MEAN
4 . BINOMIAL DISTRIBUTION
5 . POISSON DISTRIBUTION
6 . NORMAL DISTRIBUTION
7 . CHI-SQUARE DISTRIBUTION
8 . CHI-SQUARE TEST
9 . STUDENT'S T-DISTRIBUTION
10 . STUDENT'S T-DISTRIBUTION TEST
11 . F-DISTRIBUTION
12 . LINEAR CORRELATION COEFFICIENT
13 . MULTIPLE LINEAR REGRESSION
14 . LINEAR REGRESSION
15 . NTH ORDER REGRESSION
16 . GEOMETRIC REGRESSION
17 . EXPONENTIAL REGRESSION
18 . MEAN-VAR-ST DEVIATION
m) missing values? unknown, not probable
2. EPISTAT 3.0/ 1984 / Tracy L. Gustafson / IBMPC & BASICA / sw $25 / basic stat
a) EPISTAT
b) 3.0
c) 1984
d) interpreted BASIC: BASICA / GWBASIC
e) English
f) MINIMUM OPTIMAL
IBM PC with 64K RAM IBM PC with 96K RAM
One 160K disk drive Two 320K disk drives
MDA Color graphics adapter
Monochrome monitor Hi-res color monitor
BASICA BASICA
IBM or Epson printer with graphics
g) PCSIG#88
h) ShareWare $25
i) Tracy L. Gustafson, M.D.
j) 1705 Gattis School Road / Round Rock, Texas 78664
k) EPISTAT is a collection of programs written in BASICA for statistical
analysis of small to medium-sized data samples ( < 28 samples or variables
and < 2000 total data entries per file). The 25 programs in EPISTAT perform
more than 40 common statistical tests or functions and provide utilities for
data entry, editing, printing, graphing, sorting, selecting, transforming and
crosstabs.
l) PROGRAM PROGRAM
TEST OR FUNCTION NAME TEST OR FUNCTION NAME
---------------- -------- ---------------- -------
Analysis of variance........ANOVA* McNemar's test.............MCNEMAR
Bayes' theorem..............BAYES Mean, Median and S.Dev.....DATA-ONE*
Binomial distribution.......BINOMIAL Normal distribution........NORMAL*
Chi-square test.............CHISQR Poisson distribution.......POISSON
Correlation coefficients....CORRELAT* Random sample generator....RANDOMIZ
Crosstab reports............XTAB* Rank sum test..............RANKTEST*
F Distribution..............ANOVA* Rates adjusted.............RATEADJ*
Fisher's exact test.........FISHERS Sample size determination..SAMPLSIZ
Graph histogram.............HISTOGRM* Select specific records....SELECT*
Graph scattergram...........SCATRGRM* Signed rank test...........RANKTEST*
Linear regression...........LNREGRES* Student's T test...........T-TEST*
Mantel-Haenszel Chi-square..MHCHISQR Transfer data between
Mantel-Haenszel for two EPISTAT files.......FILETRAN*
multiple controls/case....MHCHIMLT* Transfer data from
FORTRAN datafiles.......FORTRANS*
*Starred programs can evaluate data entered and saved using DATA-ONE.
m) Determination of the most appropriate statistical test;
missing values support; statistical reference list.
3. SPPCdemo 1.0 / 1984 / WALMYR Publishing Co. / IBMPC / sw $75 / statistics
a) SPPC (Statistical Package for the Personal Computer) - limited demo version
b) 1.0 {& 3.0}
c) 1984 & ?
d) C (Lattice C 2.10)
e) English
f) IBMPC & ANSI.SYS
g) IBM-PCNL#135 / PCSIG#232 / CP/MS2232 {& PCSIG#623}
h) sw $110 for full SPPC; $25 commission in return each time when someone else
registeres your registered copy and/or purchases modules for > $75.
i) WALMYR Publishing Company
j) PO Box 3554, Leon Station, Tallahassee, FL 32315
k) The SPPC is a fully interactive statistical program with which one may enter,
manage (edit, sort/reorder, select), and analyze simple and complex sets of
data on file. It contains simple descriptive statistics, a wide range of
parametric and non-parametric tests, several types of simple and complex
regression analysis, several types of factor analysis, [and more?].
Not available in this demo version are data entry and managing facilities.
l) Statistical techniques
1) Descriptive statistics and additional procedures
a> (Univariate) summary statistics with up to 50 variables from raw data input
(keyboard or diskfile) records having 200 or fewer variables: the arithmetic
mean, biased and unbiased standard deviations, skewness, kurtosis, minimum
and maximum values, the range, and the geometric and harmonic means.
b> Frequency distributions (not in demo version)
c> Contingency tables (not in demo version)
2) Basic probability statistics or related functions. Input from keyboard.
a> Factorials
b> Combinations & permutations
c> Poisson probabilities
d> Binomial probabilities
e> Geometric probabilities
f> Hypergeometric probabilities
g> Exponential probabilities
h> Ordinary chi-square probabilities
i> Normal curve probabilities
j> t-ratio probabilities
k> f-ratio probabilities
3) Measures of association
a> Pearson correlation matrix from (a selection of) 200 or fewer variables in a
raw data file with an unlimited number of cases: means, standard deviations,
and correlations for 51 or fewer variables. Ouput to disk file.
b> Non parametric association (not in demo version)
4) Parametric tests (not in demo version)
5) Non-parametric tests (not in demo version)
6) Analysis of variance (not in demo version)
7) Regression analysis: both simultaneous and hierarchical multiple regression
analysis using up to 50 independent variables from (a selection of) a raw
data file (with max. 200 variables) or a summary data file with a maximum of
51 variables. Output to screen, printer and/or disk file.
8) Test construction (not in demo version)
9) Calculator (Reverse Polish notation)
m) Missing value codes are supported as any single numerical value.
Three types of data entry are available:
1) Data from keyboard for some procedures,
2) Raw-score (ASCII) data files (not rectangular matrices, but some kind of a
space delimited data file,
3) Summary data files, a specially formatted blank delimited ascii file,
containing sample size, NofVars, variable names, means, standard deviations,
and correlations.
4. KWIKSTAT 1.0 / 1986 / Alan C. Elliott / IBMPC + CGA / sw $35 / statistics
a) KWIKSTAT
b) 1.0
c) 1983 / 1986
d) compiled BASIC (MS BASIC 5.60)
e) English
f) 1. An IBM PC, XT, AT or compatible computer, preferably with a hard disk.
2. DOS 2.0 or higher
3. At least 196K of memory
4. A graphics (CGA), EGA or PGA monitor (displaying CGA). No Hercules.
g) PCSIG#654 & PCSIG#655 (2 disks)
h) sw $35 (incl. bound manual); extra manual $10; update $10; shipping $4
i) Alan C. Elliott
j) Mission Technologies, P.O. Box 1169, Cedar Hill, Texas 75104; (214)-291-2115
k) KWIKSTAT is a graphics oriented program for scientific statistical analysis.
It is entirely menu controlled. It allows for a maximum of 128 fields
(variables).
l) Statistical and report procedures. The basic divisions of KWIKSTAT:
1) DATA: Create a KWIKSTAT-database, enter and edit data from:
a> Keyboard (variable names, max. field width, Character/Numeric/Date/Logical);
b> ASCII file (raw data);
c> dBASE file (.DBF) (transformations, new variables, grouping variables, etc.
in dBASE).
2) REPORT: Output reports about the data, output ASCII text files, and examine
the contents of the database or other files.
a> Output of a formatted report: listing of the data in the dataset (or a
selected subset of the database) with titles, column headings and optional
record numbers.
b> Output of the raw data into a standard ASCII TEXT file for transferring data
to other programs. Optional output of a "format" file, which describes the
contents of the text file in dBASE- or in SAS-format (with extension .SAS).
3) STAT1: Descriptive statistics & graphical presentations:
a> Descriptive statistics:
1> Detailed statistics on a single variable, a.o. sample size, mean, standard
deviation, minimum, and maximum, and SEM.
2> Summary statistics on a number of variables.
3> Crosstabs (contingency tables) and Chi-square or where appropriate
Fisher's exact test and Yates Chi-Square.
4> P-value determination for Z, t, F and chi2 (2-sided).
b> Graphical presentations:
1> HISTOGRAM : distribution of a continuous variable
2> SCATTERPLOT : relationship between two variables
3> BOX PLOT : 0, 25th, 50th, 75, and 100 percentile
4> 3-D BAR CHART: three dimensions of a table of counts (enter data in
contingency table procedure). Also view 'slices'.
4) STAT2: Comparative statistics:
a> Student's t-tests (both independent and paired), 2 groups.
b> One Way Analysis of Variance (ANOVA) on 3-10 independent groups or 3-10
repeated (paired) observations.
c> Newman-Keuls multiple comparisons on the 0.05 signifiance level.
5) STAT3: Simple and Multiple Linear Regression (up to 10 independent variables)
a> Comparing paired observations, simple or multiple linear models
b> t-test for significance of parameters, slope, R2, coefficients and intercept
c> Pearson correlation coefficient, R2, also adjusted
d> plot of residuals from linear regression fit
6) STAT4: Survival Analysis, Life Tables, comparison of survival rates
a> a (cumulative) plot of survival times
b> If more than one group is entered, a Mantel-Haenszel test is performed to
test the hypothesis of equal survival patterns for each group.
m) sample data sets, statistical reference list. {NO MISSING VALUES SUPPORT}
5. SuperSTat (SST) 1.00 / 1986 / Mike Derby / IBMPC / sw $35/$20 / market stats
a) SuperSTat (SST)
b) 1.00
c) 1986
d) compiled BASIC (MS BASIC 5.60)
e) English
f) IBM PC/XT/AT with MS/PC/DOS 2.0 or higher, 256 Kb memory and any monitor
(text mode graphics, IBM graphics characters), a hard disk or two disk drives
and an EPSON compatible (or other) printer.
g) PCSIG#638 & PCSIG#639 (2 disks)
h) sw $35; students: $20; full commercial version available!
i) Mike Derby; Derby Micro-Computer Services, Inc.
j) P.O. Box 220566, Charlotte, NC 28222, (704) 536-7721
k) Superstat Share is a shareware Statistical Package for analyzing the results
of Market surveys and other questioneers (poles). It allows weighting the
data set, encryption of data and automating analysis by using special batch
file commands (piping). The program is menu controlled (filling in forms on
the screen). This version allows 25 variables and 50 data columns with up to
nine values to the variable. (The commercial version allows 250 variables and
250 data columns with up to 9 values each.) Both versions handle an unlimited
number of records. {'Data columns' indicates maximum input line length.}
l) Data specification and statistical procedures:
1) data:
a> use of separate CODEBOOK files to set the number of columns and keep track of
maximum and minimum values (with interval level variables). Editing and
extending facilities. Titles, labels, levels, columns, maximum and minimum.
b> entry: keyboard or diskfile
c> data reformat: joining/merging/splitting/encrypting data files
d> data manipulation, recoding: permanent/temporary
e> selection based on values, logical expressions
f> weighting
g> variable types:
1> nominal
2> ordinal
3> interval (1-12 digits, data columns)
4> multiple response (MRV); tabulation
I: MSA: a stand alone multiple response variable
II: MTO: Tab-of-other MRV: a multiple tabulation of a single value of other
variables. A MRV may have up to nine values and use up to ten data columns
with its required missing value switch.
2) statistical procedures:
a> FREQ: frequency distributions: Bar chart, Value counts, Value percentages,
Mean Ordinal Rank, Mode, Median, Standard Error at 95% confidence level.
b> CROSS: crosstabs; bar chart (line printer text, no graphics)
1> Cell, row, column, and total counts and percentages and missing values.
2> Measures of association (and significance levels):
I: nominal to nominal: Goodman and Kruskals Tau and Chi-Square
II: ordinal to ordinal: Gamma and Z-Score Gamma
III: nominal to ordinal: Wilcoxons Theta and contingency coefficient
3> Cramer's V, the degrees of freedom, N, and missing values.
c> SCATTER: scatterplots, interval- & ratio-variables
Stats for bi-variate correlation, regression, and one-way ANOVA;
descriptive stats on the individual variables.
m) sample data sets, missing data support, statistical reference list
6. RA 1.5 / Jan. 1987 / Mark E. Williams / IBMPC / sw $45 / simple statistics
a) Research Assistant
b) 1.5
c) January 1987
d) compiled BASIC (IBM)
e) English
f) IBMPC + CGA (Epson FX-80)
g) ?
h) sw $45 ($65 also mentioned)
i) Mark Williams
j) 616 Beechtree Court, Chapel Hill, N.C. 27514
k) Basic statistics package, divided into separate programs
l) MODULE 1 MODULE 2 MODULE 3 MODULE 4
DATA MANAGEMENT STATISTICS STATISTICS STATISTICS
Enter Data Scattergram 2 X 2 Tables 1 sample T-test
Store Data Frequencies Simple Regression 2 sample T-test
Edit Data Descriptive Stats Multiple Regression Paired T-test
Print Data Cross Tabs Correlation One-way ANOVA
m) limitation to 25 variables and 200 records
variable types (continuous/categorical), conversion between types;
both variable and value (category) labels;
max. nr. of categories is 10, but this number sometimes may be exceeded (?);
very user friendly: all variable names presented for choice with cursor keys;
transformations (logarithmic and exponential), but no conditional selections;
read .DIF files.
7. STATistiek / 1987 / Piet van Blokland / IBMPC / sw f50 / enquete & statistiek
a) STAT(istiek)
b) initial release
c) 20 september 1987
d) (TURBO) PASCAL
e) Nederlands (Dutch)
f) NIVO-apparatuur: IBMPC (+ CGA; andere GA: bel 020-5601231), 512Kb
g) IBM-PCNL630 (software ook te copieren bij de Regionale Steunpunten Onderwijs
en Informatietechnologie (of tegen behandelingskosten bestellen bij de VL.)
h) sw f50
i) Piet van Blokland, Hogeschool Holland, Vrije Leergangen VU
j) VL vakgroep wiskunde/informatica, postbus 261, 1110 AG DIEMEN, 020-5601911 //
Faculteit Wiskunde & Informatica, Hogeschool Holland, De Boelelaan 1081,
1081 HV Amsterdam.
k) enquete-ontwerp, -invoer en -verwerking (voor gebruik op scholen)
l) inhoud (uit menu's):
1) Enquete opzetten (invoeren, wijzigen etc.)
a> o.a. verificatie op minimale en maximale waarden
2) Gegevens proefpersonen invoeren
3) Statistische verwerking van de gegevens
a> Data manipulatie
b> frekwentie tabellen
c> histogrammen (cumulatief)
d> gemiddelde,spreiding,mediaan,boxplot
e> kruistabellen (chi-kwadraat-toets)
f> Correlatie matrix
g> lineaire regressie & scatterdiagram
h> Multipele Regressie
i> Lineaire discriminant analyse
j> Wilcoxon rangsom toets en de t-toets voor verschillen
k> Wilcoxon rangteken toets en de t-toets voor paarsgewijze verschillen
l> Kendall`s tau en Spearman`s rho
m> Lorentz curve
n> cirkel diagram
o> tijdreeksanalyse (additief / multiplicatief)
m) Extra mogelijkheden, o.a.:
1) Enquete veranderen
2) Transformatie uitvoeren
3) De proefpersonen op volgorde van hun nummer wegschrijven
4) Deelgroep maken
5) Twee enquetes koppelen
6) Omzetten bestanden voor andere programma's
7) Omzetten bestanden van andere programma's voor STAT
8. TSxSTyy / 27 Nov.1988 / Timo Salmi / IBMPC&CGA /pd/ statistical programs
a) TSxSTyy, in which x=1..4 and yy=version number; four separate programs:
(1) STATMEAS (2) STATREGR (3) STATTRAN (4) STATRCOR
b) (1) TS1ST14 (2) TS2ST14 (3) TS3ST12 (4) TS4ST12
c) 27 November 1988
d) Turbo Pascal
e) English
f) IBMPC (& CGA optional for STATREGR only) (at least 256k?)
g) Simtel-20 PD archives at WSMR-SIMTEL20.ARMY.MIL (ARPA-Internet) or its peer
fileservers at LISTSERV@NDSUVM1 / LISTSERV@RPIECS / TRICKLE@TREARN;
BBS's: 2:515/1 (Micro Maniacs Opus in Finland, tel. 358-0-425966).
Internet (NorduNet or FuNet) 128.214.12.3 by anonymous ftp.
More info on file request and ftp usage from ts@chyde.uwasa.fi.
Also appearing as UseNet Binary Postings in comp.binaries.ibm.pc.
h) pd (a registered version exists, with larger capacities, price unknown)
i) Timo Salmi; Professor of Accounting and Business Finance
j) School of Business Studies, University of Vaasa
P.O. BOX 297, SF-65101 Vaasa, Finland
Internet address: ts@chyde.uwasa.fi
Funet address: VAKK::SALMI
FidoNet address: 2:515/1 (Micro Maniacs Opus, To: Timo Salmi)
Bitnet address: SALMI@FINFUN
k) All programs include a built-in help system, which can be invoked by typing ?
at any interactive question.
1) STATistical MEASures program is intended for univariate statistical analysis.
2) STATistics: multiple REGRession analysis.
3) STATistics: TRANsformations (also keyboard input, but no real editing).
4) STATistics: Ranks and CORrelations.
l) individual features:
1) STATMEAS calculates for a set of observations:
a> Standard statistical measures
b> Autocorrelations
c> Runs
d> Draws frequency diagrams
2) STATREGR also draws (or rather writes) low-resolution (text, to screen,
printer or diskfile) or high resolution (CGA, to screen only) scatter
diagrams of the data, and the regression analysis.
Also calculation of ABS DEVIATION FROM UNIT MATRIX. The smaller this figure,
the less probability of computationally weak estimates.
3) The following transformations are available from STATTRAN:
a> LOGAE is ln(x(i)), i.e. takes the natural logarithm.
b> LOGAR is log(x(i)), i.e. takes the logarithm in a given base.
c> POWER is x(i)^a, i.e. raises to the given power.
d> SCALE is a*x(i), i.e. scaling with a constant coefficient.
e> MULTI is y(i)*x(i), i.e. multiplies a variable with another variable.
f> DEFLA is y(n)/y(i)*x(i), i.e. deflates a variable with another variable.
g> DIFFE is x(i)-x(i-1), i.e. takes differences.
h> LAGLD is x(i+lag), i.e. a lag or a lead.
4) STATRCOR optionally calculates:
a> The ordinary Pearson's correlation coefficients;
b> The t-test values of Pearson's correlation coefficients for testing their
statistical significance;
c> Spearman''s rank correlation coefficients;
d> Ranks of the observations for any individual variable.
m) The data can either be given from the keyboard or taken from a file.
The maximum number of variables is 25. The maximum number of observations is
400 (for each variable). The public domain version, however, sets the limits
at 4 and 30 respectively. Data format is blank or comma delimited records.
Missing value support.
9. MAPSTAT 2.0 / 1988 / Douglas L. Anderton / IBMPC / pd / statistical package
a) MAPSTAT, Multivariate Analysis Package
b) 2.0 (7th release)
c) 1985-88
d) Turbo Pascal vs. 2/3
e) English
f) IBMPC, 56k (!) (small system, e.g. floppy only; also under multitasking)
g) Simtel-20 PD archives at WSMR-SIMTEL20.ARMY.MIL (ARPA-Internet) or its peer
fileservers at LISTSERV@NDSUVM1 / LISTSERV@RPIECS / TRICKLE@TREARN | author;
h) pd (even free if registered!)
i) Douglas L. Anderton
j) Department of Sociology
University of Chicago
1126 E. 59th Street
Chicago, IL 60637
k) MAPSTAT is a very serious multivariate statistical analysis package capable
of meeting 90% or more of most users' analytical needs. It provides the most
frequently used procedures of the relatively expensive statistical packages.
MAPSTAT can theoretically analyze an unlimited number of variables and cases.
It has been tested on data files with over 200 variables and 10,000 cases.
Blank delimited numeric data. (Conversion of fixed to free formatted data.)
Input from keyboard or file. Output to screen, printer or file.
l) The currently available subprograms in MAPSTAT include (with present limits):
1) TRANSFRM - data transformations, selection and recoding; max. 100 variables,
incl. created; Transformation language in Reverse Polish Notation.
Valid Arithmetic Operators: + - * /
Turbo Pascal Functions Supported:
ABS ARCTANN COS EXP FRAC INT
LN SIN SQR SQRT ROUND TRUNC
RANDOM
Nonstandard MAP functions supported:
CASEN IF IFS LAG MOD NORMAL POW REC
Number Entry:
Leading minus allowed (not plus) number must be less than or
equal to 11 digits, e.g. .001 12 -.0000005 etc.
2) DESCRPT - descriptive statistics and frequency histograms; max. 100 var's;
3) CORREL - correlation and covariance matrices; max. 50 variables;
Output of CORREL may be used as input for REGRESS and other procedures.
4) REGRESS - multiple linear regression; max. 30 variables;
Calculation of residuals via TRANSFRM, plotted with PLOT.
Coding of dummy variables by TRANSFRM for use in calculation of complicated
general linear model analysis of variance (e.g. GLM/ANOVA's) through REGRESS.
5) CROSSTAB - n-way crosstabulation and association tests; Association measures.
Finite number of real values. Max. 7 dimensions, 8 variables, 25
values/variable (categories), 3500 cells.
6) HYPOTHS - simple hypotheses test on means and variances; max. 2 variables
incl. group variables
7) PARTIAL - partial correlation and covariance coefficients; max. 30 variables
8) FACTOR - principle axis factoring with orthogonal rotations; max. 30 var's;
Calculation of factor scores with TRANSFRM using the score coefficients.
9) CLUSTER - clustering by kmeans algorithm; max. 10 variables, 25 clusters;
10) PLOT - simple 2 dimensional plots or scattergrams; max. 30 variables;
11) MANOVA - multiple dependent variable analysis of variance; max. 30 var's;
12) FIXFREE - utility for fixed to free format file conversion; max. 255 var's;
13) SORT - utility to sort mapstat files with DOS SORT;
m) Optional support of codebook files with variable names and missing values.
Weight variables (DESCRIPT and CORREL).
B. Single-purpose, separate statistical programs:
1. GRAPH2 2.0 / May 20 1984 / Bob Shannon / IBMPC + CGA & BASICA / pd $20 / plot
a) GRAPH2
b) 2.0
c) May 20 1984
d) interpreted BASIC: BASICA / GWBASIC
e) English
f) IBM PC
Color graphics card
Color TV or monitor display
(Monochrome version available for Hercules Graphics Card)
One disk drive (single or double sided)
64k Memory
DOS 2.0
IBM printer or compatible for printout
g) ?
h) pd $20
i) Bob Shannon
j) 228 Tulare St. / Brisbane, California 94005
k) GRAPH2 is a BASIC program to plot line_ drawing type figures on the graphics
screen of the IBM-PC computer.
l) The program allows arbitrary specification of the X and Y axes on the figure,
linear or logarithmic axes, and labeling. The data and/or the specifications
of the plot are entered from the keyboard or disk file, may be edited
interactively and restored. Up to five (5) sets of data can be plotted on the
same figure. Each set may have its own symbol type, symbol size, and type of
line connecting the points.
m) plot linear regression line
2. SPLINE 2.01 / October 1986 / D.A.A.N. / IBMPC+CGA & GWBASIC / smooth curve
a) SPLINE
b) 2.01
c) October 1986
d) C (MS C 3.0) (and interpreted BASIC)
e) (English/Dutch)
f) IBMPC + CGA + GWBASIC
g) -
h) pd
i) D.A.A.N. (?)
j)
k) Interpolate Smooth Curve
Spline takes pairs of numbers from the standard input as abscissae and
ordinates of a function. (A minimum of four pairs is required.) It produces a
similar set, which is approximately equally spaced and includes the input
set, on the standard output. The cubic spline output has two continuous
derivatives and a sufficient number of points to look smoothly when plotted.
l) The following options are available:
1) Supply abscissae automatically
2) Influence boundary value computation
3) Specify number of intervals, default spacing is 100 intervals
4) Make output periodic
5) Specify lower and upper limits
m) additional utility programs:
1) PLOTTER(2) to (line)plot both raw and smoothed data
2) RANDOM(2) to generate random raw data
3. CURVEFIT 1.00 / May 21 1984 / Thomas S. Cox / IBMPC / pd / curve fitting
a) CURVEFIT
b) 1.00
c) May 21 1984
d) interpreted BASIC: BASICA / GWBASIC
e) English
f) IBMPC
g) ?
h) pd -
i) Thomas S. Cox
j) 102 Evergreen Street Easley, SC 29640
k) Curve fitting for positive values X and Y (presently up to 100).
Data may be edited and stored on file.
l) Fitted equations:
1. Y=A+B*X STR. LINE 2. Y=B*X LINE THRU ORG.
3. Y=1/(A+B*X) REC. STR LINE 4. Y=A+B*X+C/X LIN AND RECIP.
5. Y=A+B/X HYPERBOLA 6. Y=X/(A*X+B) RECIP HYPERBOLA
7. Y=A+B/X+C/X*X 2ND ORD HYP 8. Y=A+B*X+C*X*X PARABOLA
9. Y=A*X+B*X*X PAR AT ORIGIN 10. Y=A*X^B POWER
11. Y=A*B^X MOD. POWER 12. Y=B^(1/X) ROOT
13. Y=A*X^(B*X) SUPER GEOMET. 14. Y=A*X^(B/X) MOD GEOMETRIC
15. Y=A*e^(B*X) EXPONENTIAL 16. Y=A*e^(B/X) MOD EXPONENTIAL
17. Y=A+B*ln(X) LOGARITHMIC 18. Y=1/(A+B*ln(X)) RECIP LOG
19. Y=A*B^X*X^C HOERL FUNCTION 20. Y=A*B^(1/X)*X^C MOD HOERL
21. Y=A*e^((X-B)/2) NORMAL 22. Y=A*e^((ln(X)-B)^2/C) LOG NORMAL
23. Y=A*X^B*(1-X)^C BETA 24. Y=A*(X/B)^C*e^(X/B) GAMMA
25. Y=1/(A*(X+B)^2+C) CAUCHY
m) coefficient of determination corrected for degrees of freedom.
4. REGRESSI // Keith D. Barber / IBMPC / pd / linear & geom. & exp. regression
a) REGRESSI
b) -
c) 1984 ?
d) interpreted BASIC: BASICA / GWBASIC
e) English
f) IBMPC
g) ?
h) pd -
i) Keith D. Barber - librarian
j) IBM PC User Group of Greater Kansas City
k) This program fits a straight line, a geometric curve, and an exponential
curve to a given set of coordinates using the method of least squares.
l) The equation of each curve,the respective coefficient of determination (R 2),
coefficient of correlation (R), and standard error of estimate are output.
The form of the three equations are:
Straight Line Geometric Curve Exponential Curve
y = mx + b y = bx^m y = bm^x
m) interpolation
5. REGRESS / 4-11-83 / Engineering Software Inc. / IBMPC / pd / curve fitting
a) REGRESS
b) -
c) 4-11-83
d) interpreted BASIC: BASICA / GWBASIC
e) English
f) IBMPC
g) PCSIG#180
h) pd -
i) Engineering Software Inc.
j) 2000 Washington St., Wilmington, Delaware (DE) 19802
k) Curve fitting program for regression analysis of a two dimensional set of
data. The REGRESS program uses regression analysis to find equations of
curves that most closely fit the ordered data sets (X,Y) entered, and then
allows the user to interpolate values based on the equation selected.
l) At least two independent points must be entered, and up to 50 points can be
accomodated. The program calculates (where possible) and prints the
coefficients (A,B,C) for equations in the following forms:
1. y = A + B*x
2. y = A + B*x + C*x*x
3. x = A + B*y
4. x = A + B*y + C*y*y
5. y = A + B*log(x)
6. y = A + B*log(x) + C*log(x)*log(x)
7. x = A + B*log(y)
8. x = A + B*log(y) + C*log(y)*log(y)
9. log(y) = A + B*log(x)
10. log(y) = A + B*log(x) + C*log(x)*log(x)
11. log(x) = A + B*log(y)
12. log(x) = A + B*log(y) + C*log(y)*log(y)
Of course, the program will not be able to evaluate equations that
contain log(x) when any of the data entered has values of x less than
zero. The same applies to y and equations that contain log(y). The
residual sum of squares is printed for each equation as [fit error].
m) interpolation and extrapolation / negative values as well
6. SORBUS 1.0 / 01-21-87 / F.J.H. Lupschen / IBMPC / pd / bar charts
a) SORBUS
b) 1.0
c) 01-21-87
d) compiled BASIC
e) English
f) IBMPC, any (text) monitor (IBM graphics characters)
g) {download from bulletin board}
h) pd
i) F.J.H Lupschen
j) The Toxic Project/DataFlex/ProToViSion/SoftHouse/ProToSoFt, Brunssum,
045-273296
k) Simple bar charts on text screen with colour and graphic characters
l) plot options (titles, tic marks, labels, grid, axes), store and recall data;
time sequence bars
m) peculiar user interface, overaccented minor aspects, not logical menu
structure, though smart and acceptable bar charts
7. STATPROG / 1986/87 / Joseph C. Hudson / IBMPC / pd $5/- / statistics progs.
a) {STATPROG}, several separate statistical programs
b) LILFOR: 1.3, PROB: 1.1, {rest: unknown}
c) 1986 / 1987
d) CFIT/LILFOR/SIMCORR: TURBOPASCAL / PROB: compiled BASIC
e) English
f) IBMPC
g) PCNL#695
h) pd CFIT: $5, LILFOR13: -, PROB: -, SIMCORR: -
i) Joseph C. Hudson
j) 4198 Warbler Dr., Flint, MI 48504
k) A number of separate statistical programs:
CFIT fits up to 196 different curves to paired data using Least Squares
Regression. LILFOR13 prints a graph that can be used to hand plot a sample
CDF to perform the Lilliefors test for normality; it also reads data from
disk or keyboard and performs the Lilliefors test. PROB.EXE computes
probabilities for the Binomial, Negative Binomial, Poisson and Hypergeometric
distribution. SIMCORR produces pseudo-ramdom samples from a bivariate normal
population.
l) programs & descriptions:
1) CFIT12 COM 43194 12-17-86 Fit curves to paired data with LSR
CFIT1287 COM 41801 12-17-86 CFIT for use with 8087 chip
CFIT fits up to 196 different curves to paired data using least squares
regression. The program can report results sorted by F value or by adjusted
coefficient of determination. It can plot the observed data together with any
of the fitted curves. CFIT can report a variety of diagnostic information,
including histograms of residuals, anova tables and various standard errors.
Fitted Y values may be calculated. The regressions generated by CFIT can be
saved to disk for future use.
Data Requirements:
CFIT accepts data in SDA, space delimited ASCII, format.
Missing data must be coded with some value, not just left out, since each row
must have the same number of entries. Any number that does not otherwise
occur in the data set can serve as the missing value code. There can be up to
250 rows of data and any number of columns. CFIT will only read the two
columns it needs, the column of X values and the column of Y values.
Some options:
Compute Y Hat Values
Print Details of a Fit
Find Y in Original Units Given Y in Transformed Units
Print Histogram Of Residuals
Print Table of Residuals
Plot Fitted Curve
Report Regression Results
2) LILFOR13 COM 27292 2-07-87 Perform Lilliefors test for normality
This program does two things. It prints a graph that can be used to hand
plot a sample CDF to perform the Lilliefors test for normality. It also reads
data from disk or keyboard and performs the Lilliefors test, printing both
graphic and numeric output.
To do the test, there must be between 4 and 255 pieces of data. (The upper
limit of 255 may be increased.) Data in disk files must be in space delimited
ASCII (SDA) format. Variables are in columns. Each case is in a separate row,
with one or more spaces separating each number. All data should be numeric;
no letters.
3) PROB111 EXE 57496 8-25-86 Compute probabilities for 4 types distr.
PROB.BAS computes probabilities for the Binomial, Negative Binomial,
Hypergeometric and Poisson distributions. It computes probabilities and
percent- age points for the Standard Normal, Student's T, Chi-Square and F
distributions. The program is menu driven.
PROB.BAS uses finite series expansions to compute T, F and Chi-Square
probabilities. A combination of MacLauren series and continued fraction
expansion is used to compute standard normal probabilities. Percentage points
for the continuous distributions are computed using a one dimensional search
with nonlinear interpolation.
1.BINOMIAL 2.NEGATIVE BINOMIAL 3.POISSON 4.HYPERGEOMETRIC
5.STD NORMAL 6.STD NORMAL INVERSE 7.STUDENT'S T 8.STUDENT'S T INVERSE
9.CHI SQUARE 10.CHI SQUARE INVERSE 11.F 12.F INVERSE
4) SIMCORR COM 46680 9-02-86 Illustrate the correlation coefficient
Simcorr produces pseudo-ramdom samples from a bivariate normal population.
The main use of this program is to illustrate the meaning of the correlation
coefficient. Simcorr can help develop a feeling for the variability in the
relationship between the population and sample correlation coefficients. The
amount of variability might suprize some.
m) statistical reference lists per program
8. STATOOLS / 1985/86 / Gerard E. Dallal / IBMPC / pd / statistics utilities
a) STATOOLS (STAT TOOLS), several separate statistical programs
b) see separate programs
c) see separate programs
d) FORTRAN (MS FORTRAN)
e) English
f) IBMPC; 256Kb memory
g) PCSIG#508 & PCSIG#509 (2 disks)
h) pd
i) Gerard E. Dallal
j) USDA Human Nutrition Research Center on Aging at Tufts University,
711 Washington Street, Boston, MA 02111 and
Tufts University School of Nutrition, 132 Curtis Street, Medford, MA 02155
k) Statistical utilities:
PC-SIZE, PC-MULTI, FORGET-IT, STAT-SAK, PC-PITMAN, PC-EMS, PC-PLAN
l) Summarizing information per program:
1) PC-SIZE, version 2.0, August 8, 1985.
PC-SIZE determines the sample size requirements for single factor experi-
ments, two factor experiments, randomized blocks designs, paired t-tests, and
other experiments. PC-SIZE can calculate the power of specific sample sizes
as well as determine the sample size needed to achieve specific power.
2) PC-MULTI, version 1.0, October 15, 1985.
PC-MULTI constructs simultaneous confidence intervals for pairwise mean
differences using Tukey's honest significant differences (Studentized range
statistic). Keyboard or file data input. Text-graphics.
3) FORGET-IT, version 1.0, November 4, 1985.
FORGET-IT produces Forget-it plots, also called two-way plots. These were
introduced by Tukey (1970, chapter 16) as a graphical technique for
representing the interaction structure in a two-way table. Prompt driven,
optional file input. Text-graphics.
4) STAT-SAK, version 1.0, February 17, 1986.
STAT-SAK, the statistician's Swiss Army Knife, is meant as a supplemental
tool for anyone who analyzes data regularly and has access to a large
statistical package. It was written under the assumption that users would
have access to a "comprehensive" micro or mainframe statistical package such
as SAS, SPSS-X, BMDP, or SYSTAT. STAT-SAK will not do calculations that
require the entry of the original observations. STAT-SAK calculates
distributions (Normal, t, chi-square, F), tests of independence/homogeneity
of proportions in two dimensional contingency tables, Mantel-Haenszel test,
McNemar's test, correlation coefficients, Bartholomew's test.
a> Distributions:
1> Normal distribution
I: statistic to probability:
: upper-tail
: lower-tail
: two-tailed
II: probability to statistic
2> t-distribution: statistic to probability
3> chi-square distribution:
I: statistic to probability
II: probability to statistic
4> F-distribution: statistic to probability
b> Tests of independence/homogeneity of proportions in two dimensional
contingency tables: Pearson chi-square statistic with Yates's continuity
correction in the case of a 2 by 2 table
c> Mantel-Haenszel test, approximate 95-% confidence interval for common odds
ratio using Fleiss (1981,eq. 10.22)
d> McNemar's test: Chi-square statistic using continuity correction
e> Correlation coefficients:
1> Confidence interval for a single correlation coefficient
2> Compare two independent correlation coefficients using Fisher's z
transformation
f> Bartholomew's test for increasing proportions (table of critical values)
5) PC-PITMAN, version 2.0, December 12, 1985.
PC-PITMAN performs exact randomization tests. It also performs Wilcoxon
signed-rank tests and Wilcoxon-Mann-Whitney U tests in the presence of an
arbitrary number of ties in the data. Allows for 100 observations. Keyboard
or external file input.
6) PC-EMS, version 1.0, September 2, 1985.
PC-EMS uses the algorithm of Cornfield and Tukey(1956) to calculate tables of
Expected Mean Squares for balanced experiments.
7) PC-PLAN, version 1.0, December 12, 1985.
PC-PLAN generates randomization plans. The resulting file can be used as
input to a word processing program for touch up and entry into a report.
Prompt driven.
m) Double precision calculations are used throughout. Reference reviews.
9. NCCLSPRE 1.83 / N. Adams / IBMPC&BASICA / pd / calculates a samples precision
This program will calculate within run, run to run, and day to day precision.
{BASIC / English} PRECISION BASED ON COLLECTED DATA / PCSIG#49/180
10. NORMAL-Z / 3-3-82 / Craig W. Uthe / IBMPC&BASICA /pd/ normal distribution `Z'
This program computes areas under the normal distribution probability curve
from the mean to a designated `Z' value. This program also can compute `Z'
values from a given area. {BASIC / English} / PCSIG#180
11. HISTGRAM (displays and prints simple histogram in text mode) / IBMPC & BASICA
/ pd / {BASIC / English}
12. PIECHART (some extension of IBM's PIECHART) / IBMPC+CGA&BASICA /pd/ {English}
13. PIE2 (about the same as PIECHART) / IBMPC+CGA & BASICA / pd / {BASIC/English}
14. BAR's (several separate programs to generate bar charts) / IBMPC+CGA&BASICA /
pd / {BASIC / English}
BAR1 shows bars (histograms) of values without any values on the axes, nor
any labels (medium resolution)
BAR2 shows bars (histograms) of values with values on the y-axis and labels
on the x-axis (medium resolution). The lowest y-values lies just above the
x-axis.
BAR3 the same as BAR2, but enabling two different values per x-label, plotted
just next to each other.
BAR4 the same as BAR2, but with 3D depth effect (medium resolution).
BAR5 shows stacked bars in medium resolution with the value y=0 just above
the x-axis.
BAR6 shows stacked bars in high resolution with the value y=0 just above the
x-axis.
15. BARCHART (simple bars, max. 4 per grouping, text mode, graphic characters) /
A.N. O'Nym U.S. / IBMPC & BASICA/GWBASIC / pd
C. Function plotting programs (based on specified functions, not on data points):
1. GRAFIEK / 1988 / Piet van Blokland e.a. / IBMPC + CGA / sw f100 / plot func's
a) GRAFIEK
b) -
c) 2 januari 1988
d) (TURBO) PASCAL
e) Nederlands (Dutch)
f) IBMPC + CGA
g) -
h) sw f100
i) Piet van Blokland, Douwe Kok en David Tall, Hogeschool Holland, Vrije
Leergangen VU
j) VL (sub)faculteit wiskunde/informatica, postbus 261, 1110 AG DIEMEN,
020-5601911 // Faculteit Wiskunde & Informatica, Hogeschool Holland, De
Boelelaan 1081, 1081 HV Amsterdam.
k) Tekenen en oplossen ingewikkelde wiskundige functies
l) 1. Zoek het functie-voorschrift
2. Grafieken tekenen
3. Uitvergroten
4. Differentieren
5. Integreren
6. Oplossen van vergelijkingen (f(x)=0 en f(x)=x)
7. Taylor polynomen
8. Blancmange
9. Definieren van eigen functies
A. Parametervoorstellingen van krommen
B. Functies van twee variabelen
C. Complexe functies
D. Differentiaal vergelijkingen
E. Tweede orde differentiaal vergelijkingen
F. Simultane differentiaal vergelijkingen
2. PLOTF(X) 0.1 / 1986 / Jim Groeneveld / IBMPC + CGA & BASICA / pd / func-plot
a) PLOTF(X)
b) 0.1
c) August 1986
d) interpreted BASIC: BASICA / GWBASIC
e) English
f) IBMPC + CGA & BASICA or GWBASIC
g) -
h) pd -
i) Jim Groeneveld
j) (new) Schoolweg 14, 8071 BC Nunspeet, NL / 03412-60413 (071-178810)
k) plots of up to 10 functions
l) Plot specification menu:
Medium/High resolution graphics draw Box, aXes or None
number of functions (1-9, 0=10) draw Grid lines or Tic marks
number of X and Y intervals / interval widths / (optionally no) labeling
m) Bsave of CGA-images
3. FPLOT /1983 / David Bush / IBMPC+CGA & GWBASIC / pd / Cartesian/Polar/3D-plot
a) FPLOT
b) -
c) 1983
d) interpreted BASIC (GWBASIC)
e) English
f) IBMPC + CGA
g) -
h) pd
i) James Fuller / David Bush
j) 71046,654
k) The program allows the plotting of a function defined at line 270 in
Cartesian, or Polar coordinates or as a 3 dimensional "Solid of Rotation".
The program also allows the screen image to be saved for later recall.
4. PCGRAF /1985 / Henning Mittelbach / IBMPC+CGA & BASICA / pd / 3D-surface plot
suggestion: 200 DEF FN Y(X) = F*COS(Y/F) + F*SIN(X/F)
5. 3D-PLOT /1986 / W. van Huissteden / IBMPC+Hercules&HBASIC / pd / surface plot
a) 3D-PLOT
b) -
c) September 1986
d) interpreted BASIC / HBASIC
e) (Dutch)
f) IBMPC + Hercules graphics
g) -
h) pd
i) Walter van Huissteden
j) Beethovenstraat 346, 7204 RN Zutphen, tel. 05750-29703
k) 3D-PLOT maakt het mogelijk om waarden van een tabel-matrix [A-tabel: 10 x 10]
grafisch zichtbaar te maken, gerelateerd aan hun (X en Y) subscript-waarden.
De grafische afbeelding is een zogenaamde "surface plot".
Het programma genereert daartoe een nieuwe tabel-matrix [D-tabel: 19 x 19]
aan de hand van de waarden van de A-tabel en de volgens de gemiddelden-
interpolatie berekende tussenliggende waarden.
De uiteindelijke grafiek bestaat uit de lijn-verbinding van iedere waarde uit
de D-tabel met de 4 naast-liggende waarden.
l) U kunt het programma op verschillende manieren gebruiken:
1) Het programma is als afzonderlijk programma te gebruiken voor het zichtbaar
maken van bijvoorbeeld wiskundige funkties.
2) Het programma is als onderdeel te gebruiken in een groter programma.
D. Statistical Pascal routines:
1. STAT-PAS / May 1985 / Philip Burns / IBMPC / pd / statistical Pascal routines
a) Turbo Pascal Statistics, Trig, Utilities
b) -
c) May, 1985
d) Pascal (Turbo Pascal - w/o 8087)
e) English
f) IBMPC
g) PCSIG#435
h) pd
i) Philip Burns
j) two Chicago BBSs: Gene Plantz's BBS: (312) 882 4227
Ron Fox's BBS: (312) 940 6496
the SMUG BBS: SMUG BBS (P. Olympia): (301) 963 5249
k) This diskette contains three separate sets of TURBO Pascal routines. All but
the last three entries pertain to the package of statistical distribution
functions. The distributions include the Beta, Log Gamma, Incomplete Gamma,
F, T, Chi-square, and Normal distributions; both the forward and inverse
functions are provided.
l) routines:
1) basic distribution and support routines:
a> ALGAMA PAS Logarithm (base E) of Gamma distribution
b> CDBETA PAS Cumulative Beta distribution
c> CDNORM PAS Cumulative normal distribution probability
d> ERF PAS Gaussian error function
e> GAMMAIN PAS Evaluate incomplete Gamma integral
f> LOGTEN PAS Base 10 logarithm
g> POWER PAS Exponentiation of real to real power
h> POWERI PAS Exponentiation of real to integer power
i> POWTEN PAS Calculate power of ten (integer powers only)
j> SIGCONST PAS Global machine dependent constants for significance routines
2) point probability routines:
a> SIGCHI PAS Significance of Chi-square distribution
b> SIGF PAS Significance of F distribution
c> SIGNORM PAS Significance of normal distribution
d> SIGT PAS Significance of T distribution
3) inverse distributions (percentage points):
a> APINVB PAS Inverse Beta distribution (Fortran/Basic/Pascal)
b> BETAINV PAS Inverse Beta distribution
c> CINV PAS Inverse Chi-square (percentage point)
d> FINV PAS Inverse central F distribution
e> NINV PAS Find percentage point of normal distribution (low accuracy)
f> NINV2 PAS Find percentage point of normal distribution (high accuracy)
g> TINV PAS Inverse central T distribution
4) demonstration routines;
5) miscellaneous routines:
a> SIGALL PAS List of statistical routines on this diskette
b> TRIGFUNC DOC Description of TRIGFUNC.PAS
c> TRIGFUNC PAS Pascal procedures for common trigonometric functions
d> UTIL PAS More than 50 utility routines
UTIL features more than 50 TURBO Pascal utility routines for such diverse
functions as screen output, frames, windows, menus, serial communications,
string manipulation, loan calculations, file I/O, and more.
m) statistical references on distributions
2. PASCALIB 1.00 / 1987 / N.D. Mackay / IBMPC / sw f25 / a.o. stat. routines
a) The PASCAL Libraries
b) 1.00
c) 1987
d) TURBO PASCAL (sources)
e) English
f) IBMPC
g) PCC030
h) sw f25
i) N.D. Mackay
j) Elzendreef 52, 3137 CD Vlaardingen, The Netherlands,
bank account 12.54.83.171, Rabo Bank
k) Collection of PASCAL routines, among which many statistical ones:
1) STAT1.LIB
The statistical routines in this library cover data screening, elementary
statistics, time series, a random generator and nonparametric routines.
Included routines:
NAME FUNCTION
TestZero Tests a matrix for zero elements
TestCond Tests a matrix for given conditions
TestBounds Tests a matrix of observations on bounds
CopySubset Copies a subset of observations to a matrix
TallyAll Computes total, average, standard deviation,
minimum and maximum of all variables in a
matrix of observations
VarTab Same for one variable, including frequencies
over given class intervals.
VarTab2 Cross-tabulates statistics for a two way
classification of tow variables
Moments Calculates first four moments
TStat Find certain T-statistics on means
RndNorm Returns a normally distributed random number
Smooth Smooth or filter a time series by weights
ExpSmooth Exponential smoothing of a time series
AutoCovar Calculates the autocovariances of a series
CrossCovar Calculates the crosscovariances of 2 series
RankVector Ranks a vector of values
TieCorrect Calculates correction factor due to ties
KendallRank Correlation test by Kendall rank coefficient
SpearmanRank Same, by Spearman rank coefficient
QTest Cochran Q-test
UTest Mann-Whitney U-test
WTest Kendall coefficient of concordance
TwoWay Friedman two-way analysis of variance
TableChiSq Computes Chi-square from a contingency table
2) STAT2.LIB
Analysis of variance, multilinear and polynomial regression, factor analysis,
discriminant analysis and canonical correlation. Included routines:
NAME FUNCTION
AnaVarData Rearranges data for analysis of variance
AnaVarCalc Performs calculus for analysis of variance
AnaVarFact Final calculations on a factorial design
DiscrMatrix Computes means and a dispersion matrix
DiscrAnal Performs discriminant analysis
PolyRegData Generates independent variables and computes
statistics for polynomial regression
PolyRegCorr Constructs intercorrelations for regression
MultLinReg Performs multiple linear regression analysis
MultStat Computes multiple statistics
Eigen2 Computes Eigenvalues and vectors of a non-
symmetric matrix of form B-inverse x A.
CanonCorr Computes canonical correlation coefficients
FactorTest Computes cumulative percentage of values
greater than or equal to a given value
FactorLoad Computes a factor matrix
FactorAnal Performs rotations of a factor matrix.
E. Specific data manipulation programs/utilities:
1. SEPARATE / 1987 / Jim Groeneveld / IBMPC / pd / data reformat/split
a) SEPARATE
b) initial release
c) July 14th, 1987
d) FORTRAN (IBM FORTRAN 1.0)
e) English
f) IBMPC
g) available from author or from bulletin boards
h) pd
i) Jim Groeneveld
j) Schoolweg 14, 8071 BC Nunspeet, Nederland, 03412-60413
NIPG TNO, P.O.Box 124, 2300 AC Leiden, NL, 071-178810
k) STATGRAPHICS can not read ASCII (fixed formatted) data files with multiple
records (more than one record per case). In order to prepare existing data
files of such a kind for STATGRAPHICS the program SEPARATE separates records
with different record numbers or any number of records without record numbers
within data files into separate data files for each set of occurring or
corresponding record numbers. These resulting data files with one record per
case may then be read independently into STATGRAPHICS one after the other.
This utility may also be of value in preparing data files for other
(statistical) programs.
l) limitation: maximum input record length: 80 bytes (max. 80 characters per
line) (the 'prehistoric' FORTRAN compiler does not allow to handle this
number flexibly)
2. COPYFIX 1.0 / 1989 / Jim Groeneveld / IBMPC / pd / data (asciitext) reformat
a) COPYFIX
b) 1.0
c) 24 July 1989
d) interpreted BASIC (GWBASIC) / compiled BASIC (QuickBasic 4.0)
e) English
f) IBMPC
g) available from author or from bulletin boards
h) pd
i) Jim Groeneveld
j) Schoolweg 14, 8071 BC Nunspeet, Nederland, 03412-60413
NIPG TNO, P.O.Box 124, 2300 AC Leiden, NL, 071-178810
k) COPYFIX converts any ASCII data or text file by adding end-of-line characters
(CR/LF) after any (without restriction) specified fixed number of characters
(columns). Original (data or text) file may contain records with unlimited
length. Already present EOL characters (CR/LF) may be included (both with and
without synchronization) or excluded. Only meaningful with fixed formatted
data. No file contents other than CRLF are changed.
l) COPYFIX can carry out 3 different tasks:
1. in/excl.: reformat an (ascii) database with fixed length records but
without embedded EOL characters into a database with equal
length records, separated by CR/LF's;
2. include : reformat an (ascii) database with (equal length) records, one
(or a few) record(s) per case, separated by CR/LF's into a
database with multiple (shorter) records per case;
3. exclude : reformat an (ascii) database with multiple (short) records per
case into a database with (less or) one (larger) record(s) per
case.
4. combination of 1. and 2. and SEPARATE by running COPYFIX twice, firstly on
the original, secondly on the first result, yielding a file
divided into 'cases' (higher order records), subdivided into
'lines' (lower order records) per case, which may be split into
multiple files using SEPARATE.
m) 'Infinite' length records (a file of one 'record' without any EOL) may also
be read and written (only limited by the amount of disk space available).
3. CODEBOOK 1.0 / 1989 / Jim Groeneveld / IBMPC / pd / data reformat/reduction
a) CODEBOOK
b) 1.0
c) 26 July 1989
d) interpreted BASIC (GWBASIC) / compiled BASIC (QuickBasic 4.0)
e) English
f) IBMPC
g) available from author or from bulletin boards
h) pd
i) Jim Groeneveld
j) Schoolweg 14, 8071 BC Nunspeet, Nederland, 03412-60413
NIPG TNO, P.O.Box 124, 2300 AC Leiden, NL, 071-178810
k) This program unformats a fixed formatted ascii data file (for STATGRAPHICS)
using a user created codebook file.
STATGRAPHICS can not read ASCII data files with records longer than 640
bytes. Next, preparing to read suited ASCII files is time consuming and user-
unfriendly: the user has to create a vector containing the information of the
position of each variable in the data file. This means a lot of initial
arithmetic and errors are difficult to correct. Besides, all fields will be
read: the vector only specifies field widths of fields that are directly
following the previous fields. The sum of all widths cannot exceed 640. So
using standard STATGRAPHICS features in this case is clumsy, time consuming
and error sensitive. This asked for the current replacing solution.
l) CODEBOOK transforms any (fixed formatted) ASCII file with one record per case
of any (unlimited) length into multiple Blank or Comma delimited (or
eventually Fixed formatted or Report) data files with STATGRAPHICS (or
generally acceptable, e.g. Lotus) variable names on the first row, each
containing a user specified number of variables. This has the advantage that
all the necessary preparation can be done within any editor, creation of a
codebook file, indicating for each variable to be transformed a.o. the field
width, the starting and ending columns and the variable name. The resulting
data files may then be read after each other into STATGRAPHICS (or used with
any appropriate program). Completely blank fields (mostly representing
missing values) in the original ASCII file may be replaced automatically by
any (user-specified) numerical or character value in the resulting output
data files. This makes reading ASCII data files much more efficient, less
sensitive to errors, much quicker and more logical and surveyable.
The resulting unformatted data file(s) contain default 58 variables because
of the STATGRAPHICS limits of 640 bytes max.line length and 10 character
variable names separated by delimiters. The output records (generally
unformatted values) are preceeded by a first line with variable names. As
many (un)formatted output files are generated as are necessary to contain the
total number of variables as a multiple of the number of variables per output
file. They are named automatically by the file name of the original formatted
data file with their sequence number as the extension.
m) (unique) limit specifications:
1) maximum number of variables: 32767 (practically unlimited), default 58
2) if not reserved enough space initially when specifying the current maximum
number of variables: optional automatic (but rather slow) adaptation to the
actually required number of variables (number of array elements) read from
the codebook file
3) maximum input record length (=number of columns per line): unlimited
4) maximum column specification (interpreted bytes/record):32767*255-1=8,355,584
(practically unlimited, practically limited by available memory in BASIC)
5) number of cases: unlimited (only hardware and software (BASIC) restrictions)