# Category : Science and Education

Archive : DA16.ZIP

Filename : DA.DOC

****************************************

* *

* The DATA ANALYST *

* version 1.6 *

****************************************

(c) Copyright 1989, All rights reserved.

by

Kin-Peng Sin

P.O. Box 9316

University MS 38677

License agreement:

This manual and the accompanying program may be copied and

distributed only in the following manner:

i) given to friends

ii) uploaded to bulletin boards

iii) included in the library of a nonprofit

computer club.

The sale of this program by any establishment either separately or

as part of a 'library' diskette is strictly forbidden.

Shareware Notice:

This is not a public domain program. If you find this program to

be of use to you, please register by sending a contribution.

Contributors who send $20 or more will receive a free upgrade of

the next release. Further upgrades for registered users will cost

$5 to cover the cost of media, postage, and handling. Your

contributions will encourage me to expand the capabilities of the

program to include extended EGA and VGA graphics support,

additional statistical functions, improved printer and plotter

support, etc.

DISCLAIMER OF WARRANTY

THIS PROGRAM AND MANUAL ARE PROVIDED "AS IS" WITHOUT WARRANTY OF

ANY KIND, EITHER EXPRESSED OR IMPLIED, INCLUDING, BUT NOT LIMITED

TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS OF

PURPOSE. THE USER OF THIS SOFTWARE AND MANUAL ASSUMES ALL RISKS.

The Data Analyst. ver 1.6

TABLE OF CONTENTS

Topic page

1. Introduction . . . . . . . . . . . . . . . . . . . . . . . 1

Description

Machine and software requirements

Acronyms used in the program

2. Using the Data Analyst . . . . . . . . . . . . . . . . . . 3

3. Data File Formats . . . . . . . . . . . . . . . . . . . . 5

4. The Graphics Mode . . . . . . . . . . . . . . . . . . . . 6

Summary of the graphics mode functions

Cursor functions

5. Using the Built-In Editor . . . . . . . . . . . . . . . . 9

6. Changing the Graph Display Options . . . . . . . . . . . . 10

7. Notes on Statistical Functions . . . . . . . . . . . . . . 11

8. Errors and Error Messages . . . . . . . . . . . . . . . . 12

More notes on labeling

9. Notes on Printing and Plotting . . . . . . . . . . . . . . 13

10. Tutorial . . . . . . . . . . . . . . . . . . . . . . . . . 14

11. User Feedback . . . . . . . . . . . . . . . . . . . . . . 17

12. General Facts . . . . . . . . . . . . . . . . . . . . . . 17

- Registration/Order Form -

The Data Analyst. ver 1.6 page 1

1. Introduction

Description

The Data Analyst is a program designed to let users analyze data

in x-y pairs by simple statistical tests and regression

techniques. Pull-down menus, a built-in editor, and a flexible

graphical interface allow the user to manipulate and view data in

a variety of ways. Features include the capability to toggle

between Log and Non-log scales 'on-the-fly' and performing

regression analysis selectively on portions of a data set. A few

popular printers and HP-GL compatible plotters are supported to

reproduce on-screen graphs.

Summary of the program's features:

- Generate regression curves

Polynomial fit (up to 9th order)

Least Squares, Log, Power, Exponential, and

Binomial Power fit.

- Perform regression analysis on all or a portion of a

data set.

- Obtain basic statistical information on a set of data.

- Determine slope of two user defined points on-screen.

- Scale data (add or multiply x,y points by a

constant/factor)

- Coordinate transformation. (Log (any base), Semi-Log,

and Cartesian system)

- Transposing x-y points.

- User defined titles and labels. Labels can be placed

anywhere in the plot.

- Built-in data editor, HP-GL compatible plotter drivers.

- Generate presentation quality plots of regression lines

with or without data points.

- Convenient graph editor.

This program is targeted towards scientists and engineers who require

the means to perform basic statistics and regression analyses on

their data. The design philosophy during program development was to

create a compact package that is both easy to learn and use. Most

functions offered in the program can be accessed with just a few

keystrokes. The simplicity of the user interface in the Data Analyst

permits users to master the program quickly. This program will not

persuade power users to part with their complex statistical packages.

However, the worker in a laboratory or even a business environment

might find this program useful in analyzing and graphing relatively

small data sets quickly and efficiently.

The Data Analyst. ver 1.6 page 2

Machine and software requirements

- IBM-PC, XT, AT, or 100% compatibles

(preferably with a numeric co-processor)

- Color Graphics Adaptor

- 512k RAM or more

- One disk drive

- DOS 2.0 or higher

- Text editor (optional but recommended)

Acronyms used in the documentation and program:

ADS - Active Data Set: the data set that any selected function in

the program will act on.

NDS - total Number of Data Sets: the total number of x-y data

sets in memory. (maximum = 5)

Lg-B - Log Base

Other facts:

Data set limit - The Data Analyst currently allows up to 500 x-y

pairs per data set.

Note: The

The Data Analyst. ver 1.6 page 3

2. Using the Data Analyst

It is imperative that the user reads all of the information given

in this manual to understand fully the capabilities of the

program. Although the manual may seem to be long, the program can

be easily mastered within a short time. Some users may find it

quicker to skip ahead to the Tutorial section and start working

with the program and then returning to this section.

The program is started from the DOS prompt:

(d):>da

The program title will appear after a few seconds. Depressing the

appropriate keys will bring the user to options on the main menu.

A pull-down menu will appear for each main option. Use the left or

right arrow keys to select different options along the bar. The

up or down arrow keys allow the selection of options within each

pull-down menu. A description of each of the functions in the

pull-down menus follow.

Bar Option - FILE

Pull-Down options

Load - load a data file, (if a file is already in

memory, you will be prompted to cancel your

pick or replace the file with a new one.)

The graph will automatically be scaled to

fit the FIRST data set initially.

Edit - edit a datafile. (after entering the last

data point, hit the

point has been accepted by the editor.

Save - saves a data file to disk.

Help - displays a help screen.

Shell - exit temporarily to the

operating system.

New - erases current data sets from memory and

resets all variables to the default settings.

Quit - quit the program.

Bar Option - STATISTICS

Pull-Down options

Regression - select the desired type of regression

analysis to be performed on data

while in the graphics mode.

Basic Stats - runs simple statistical tests on

selected data set.

The Data Analyst. ver 1.6 page 4

Bar Option - GRAPH

Pull-Down options

View - enter graphics mode.

Options - modifies scaling (Cartesian and Log),

labels, additive (cx,cy), and

multiplicative (mx,my) factors.

Defaults - displays the current settings to be

used in the graphics mode.

Bar Option - PRINT

Pull-Down options

DMPrinter - print graph on dot-matrix printer.

Plotter - print graph on pen plotter.

Options - configures the type of output on the

dot-matrix printer.

Bar Option - CONFIGURE

Pull-Down options

Log Drive - lets user set drive containing data

files.

DMPrinter - select appropriate dot matrix printer

which matches user's hardware settings.

Plotter - select pen plotter corresponding to

user's hardware settings.

Com Port - select communication port to pen

plotter.

Settings - shows current system settings.

The Data Analyst. ver 1.6 page 5

3. Data File Formats

Three file formats are used. The first line of data in all three

formats must contain the following information:

[file code], [number of data sets]

A file code of 1 is used when the number of x-y pairs in each data

set is the same. Set the file code to 2 when the numbers are not

equal.

example 1: [Format 1] 3 data sets, each containing 5 x-y pairs

1,3 <- file code = 1, 3 data sets

x11,y11,x21,y21,x31,y31

x12,y12,x22,y22,x32,y32

x13,y13,x23,y23,x33,y33

x14,y14,x24,y24,x34,y34

x15,y15,x25,y25,x35,y35

example 2: [Format 2] 2 data sets, data set #1 has 2 x-y pairs

and data set #2 has 3 x-y pairs.

2,2 <- file code = 2, 2 data sets

2,3 <- line indicating number of points in each set.

x11,y11 In this case, data set #1 has 2 points and

x12,y12 data set #2 has 3 points.

x21,y21

x22,y22

x23,y23

example 3: [Format 3] 3 data sets, each containing 5 x-y pairs.

The x value of each x-y pair is the same as the x

values of the corresponding x-y pairs in other data

sets.

3,3 <- file code = 3, 3 data sets

x11,y11,y21,y31

x12,y12,y22,y32

x13,y13,y23,y33

x14,y14,y24,y34

x15,y15,y25,y35

Terminators may either be commas or spaces.

Notation :

x1- and y1- are data points in the first data set and

x2- and y2- are data points in the second data set, etc.

Note : If the built-in editor is used for data entry, the above

guidelines are followed automatically when the file is

saved. Files of format type 3 read by the Data Analyst

are saved as format type 2. Files created in spreadsheets

or other programs with the above formats and saved as text

(ASCII) files will also work.

The Data Analyst. ver 1.6 page 6

4. The Graphics Mode

Selecting the VIEW option in the GRAPH menu will invoke the

graphics mode. In this mode, a graph will be displayed according

to the specifications shown in the DEFAULTS selection of the GRAPH

menu. On the right side of the graph are five status boxes

indicating the various options that are in effect. The boxes

labeled X-Log and Y-Log show if either of the X or Y axis is in

the Log mode. A box labeled Lg-B displays the current Log base.

Below it are two boxes (NDS and ADS) indicating the number of data

sets in memory and the current active data set. The graph title

is not displayed in this mode (except in the preview option).

Situated below the graph are two text lines. The upper box

comprise (from left to right) the X-Y coordinate of the cursor

location, a message box, and the current user-selected regression

mode.

Functions that can be performed in the graphics mode:

* Display the help screen. This is accomplished by

depressing the F1 key.

* Expand the graph to cover the entire screen. [F2 key]

Use this to preview the finished graph.

* Insert labels anywhere within the graph. To do this, use

the arrow keys to direct the small cursor located near the

bottom of the graph to the desired location. Depress the

F3 key to enable text input. Hitting the

terminate the option. The program allows a maximum of 15

on-graph labels. The string starts at the point of the

cursor when F3 was pressed in the final output. Labels are

treated as special data points by the program with its

starting location defined at the point of the cursor when

F3 is pressed. The label itself is treated as a data point

symbol. If the user makes an error while typing the label,

simply hit

* Toggle between Log and Non-log scales. Use the F4, X, or Y

key to do this. Should any data set contain negative values,

the function is disabled.

* Display or Hide the grid. [F5 key]

* Display results of regression analysis. [F6] A text screen

with the appropriate data will be displayed. Note that the

user must have performed the analysis first before this

screen can be displayed.

* Determine the slope or gradient between two user-specified

points. The two points are set by pressing the F7 key when

the cursor is at the desired locations.

The Data Analyst. ver 1.6 page 7

* Perform regression analysis selectively on portions of a

data set. A segment of the data set must be determined

before this can be achieved (a segment is defined as two

points along the X-axis). This is done by first moving the

cursor to one location and hitting the F8 key. Specify the

end of the segment by repeating the above procedure. A

regression line should appear after a few seconds. Hit the

F6 key to view the results. In general, a minimum of TWO

data points must be contained within the segment to enable

the analysis. REMEMBER: The analysis is performed on the

Active Data Set (ADS). Nothing will appear to happen if

the ADS is not within the limits of the current graph

boundaries. 2nd REMINDER: Rescale the graph to fit the ADS

with the [S] key. Select the ADS using the [I] key.

* Obtain a raw screen dump of the graph. [F9]

GRAPHICS.COM must be installed first.

* Alter the speed and accuracy of the cursor. [F10]

* Change the Active Data Set. Press the [I] key until the

desired data set number is shown on the right side of the

screen.

* Scale the graph to fit the Active Data Set. [S]

* Scale graph to fit all data sets. [A]

* Perform regression analysis on the entire Active Data Set.

[F]. To view the regression line make sure the ADS is

within the current graph limits.

* Delete the most recently entered user-defined label. [C]

* Redraw the graph. [R] Use this when the graph becomes

cluttered with regression lines, labels, or gradient lines.

* Toggle between decimal and exponential tic labels. [E]

This function becomes useful in the event that the length

of the tic mark labels become excessively long. This

function is automatically used if the magnitude of the

numbers become large. Always preview with the F2 key to

ensure that all labels are properly displayed. If they

overlap or extend beyond screen limits, reduce the number

of tic marks and use exponential labels.

* Redraw the most recently computed regression line for each

data set. [P]

* Erase all of the computed regression lines from memory. [U]

* Display the legend. [L] This will not show up in the

preview mode or in the finished graph.

The Data Analyst. ver 1.6 page 8

Summary of the graphics mode functions:

F1 - displays a help screen

F2 - expands plot to full-screen

F3 - inserts on-graph label at current cursor position

F4 - toggles between Log and Non-Log scales

F5 - hide or show grid

F6 - displays regression results of most recent analysis

F7 - determine the gradient between two points on the graph

(use the cursor to pick desired points)

F8 - pick portion of data set to perform regression analysis

(use the cursor to pick desired points)

F9 - screen dump of plot (GRAPHICS.COM must be installed)

F10 - slows down or speeds up cursor

Spacebar - cycles thru regression options (ver 1.2 upwards)

[I] - change the active data set

[C] - clears most recently entered label

[R] - refreshes the graphics screen

[S] - scales graph to fit selected data set

[A] - scales graph to fit all data sets

[F] - performs regression analysis on ALL points of

the selected data set

[X] - toggles between Log and Non-Log scales on the X axis

[Y] - toggles between Log and Non-Log scales on the Y axis

[L] - display the legend

[P] - Redisplay regression lines

[U] - Remove regression lines

[E] - toggles between decimal and exponential labels

[T] - transposes the x and y values of the active data set

(make sure that any regression lines drawn prior

to the transpose operation are removed using the U

key)

Arrow keys - move cursor

Functions of the Cursor.

The cursor is a multipurpose tool used to do the following:

- determine the starting location of a label.

- specify the starting and ending points of a segment of

a data set for selective regression analysis.

- specify two points on the graph for determining the

slope.

- determine more precisely the X-Y coordinate of a data

point on the graph.

The Data Analyst. ver 1.6 page 9

5. Using the Built-In Editor

Selecting EDIT from the FILE menu will bring up the editor. If

no data is present in memory, the user will be prompted to edit

the 1st data set. If data is present, then a second menu appears

letting the user select the data set to be edited. When there is

one data set already in memory, the user can edit either data set

#1 or #2. If two data sets are in memory, data sets #1, #2, and

#3 can be edited. In short, when data is present, the user can

add additional data sets by editing the data set designated one

number higher than the total number of data sets in memory.

The editor can only be used to edit numbers within the range

-99999 to 999999. It will not be activated when data points

exceeding these limits are detected. User's wishing to

analyze data outside these limits will have to use a more

versatile text editor to enter data.

The editor functions as follows:-

* The editor starts up in insert mode. Moving the cursor

using the left or right arrow keys will cause it to

switch to the overwrite mode.

* Alt-X will delete the current hi-lighted data point.

* Alt-C will zero the current entry.

* Because the data set is automatically sorted by X-values

upon exit, no Insert feature is necessary.

* Due to the way in which the editor detects the number of

data points present, if the last data point entered is

[0,0], it will not be recognized as valid data.

However, a [0,0] point entered in the midst of other

nonzero points is recognized.

Note: A data set must consist of at least two points.

The Data Analyst. ver 1.6 page 10

6. Changing the Graph Display Options

Selecting OPTIONS from the GRAPH menu will enable the user to

change various display options. The options that can be altered

are:

* Main title, X and Y axis labels

* Number of tic marks on the X and Y axis in the cartesian

mode. (Maximum of 10 tics, minimum of 1 tic)

(default = 6)

* Display or hide the data point symbols (OFF/ON)

(default = ON)

* Connect the data points with straight lines (OFF/ON)

(default = OFF)

* Manually scale the X and Y axis. The maximum and

minimum allowable user-specified values at present are

999999 and -99999 respectively. Users are limited to

using the autoscale feature when data values exceed the

above limits.

* Change the Log base. (default = 10)

* Change the starting Log cycle. (default = -1)

* Change the number of Log cycles in the X and Y axis.

The maximum number of cycles allowed is five.

(default = 4)

* Multiply each x and y point in a data set by a factor.

(MX, MY) (default = 1, limits : -999 to 9999)

* Add each x and y point in a data set by a constant. (CX,

CY) (default = 0, limits : -999 to 9999)

Note: For some of the options, the user is forced to enter valid

data before being allowed to leave the field. (e.g. if the

minimum x limit is set to be larger than the maximum x-axis

limit, the program will sound a tone and prompt for a valid

entry)

The Data Analyst. ver 1.6 page 11

7. Notes on Statistical Functions

- The acronyms used in the window containing basic

statistical information on a data set are as follows:

STD X - standard deviation of the x values in the set

STD Y - standard deviation of the y values in the set

min X - minimum x value in the set

max X - maximum x value in the set

min Y - minimum y value in the set

max Y - maximum y value in the set

- The results displayed in the Basic Statistics

window has taken in account any multiplicative or

cumulative factors modified in the OPTIONS

selection of the GRAPH menu.

- When using the Binomial Power fit, users have the option

to enter the desired power values.

- When a segment of a data defined by the user contains

less than the required number of data points for a

polynomial fit, the message [Insufficient points] will

be shown in the message box.

- In certain cases, some data sets contain data that will

result in the creation of an ill-conditioned matrix when

users attempt to fit a polynomial curve to the data. In

such cases, the message [Singular matrix!] will be

displayed briefly in the message box. The resulting

regression line (if any) and regression results will be

suspect and should be ignored.

- A data set (or a data segment) must contain a minimum of

two points for any type of regression analysis to take

place.

The Data Analyst. ver 1.6 page 12

8. Errors and Error Messages

Error messages might appear when an error occurs in any of the

three modes:

[1] Loading or saving files. The message [File I/O error]

will appear under these conditions:

i. when the user attempts to load a file which is not a

recognized data file.

ii. disk error. write protected diskette, open disk

drive door, etc.

iii. when the user attempts to save a file to a

nonexistent directory.

[2] Running BASIC STATS on a bad data set. The message

[Warning! Illegal data] will appear if a data set

contains insufficient data points or data which is

inappropriate for analysis.

[3] When the user is in the graphics mode, the most common

error occurs (after rescaling the graph to fit a

different data set) when the user-defined labels are too

lengthy to fit within the limits of the graph. The

offending labels will be partially displayed or be seen

to overlap areas outside the graph. The solution is to

scale the graph manually to accommodate the long label.

Labels are treated as long data point symbols. An

alternative solution to this problem is not to rescale

the graph after labels have been entered.

More notes on labeling:

The label mode cannot be activated if the cursor is too

close to the right and bottom boundary of the graph.

This mode is automatically deactivated before any text

overwrites the right edge of the graph. Be wary of

labels which have been defined to start at a negative x

or y coordinate value. Switching to the log mode will

cause these labels to be erratically placed.

The Data Analyst. ver 1.6 page 13

9. Notes on Printing and Plotting

- The HP7475A and the HP ColorPro should be set at

9600,S,7,1

- For the HP ColorPro to work, it must contain the

graphics module.

- The HP Draftmaster-I should be set at 9600,N,8,1

The same applies to the Draftpro EXL.

- Wrong communications settings will result in garbled

plotter output.

- Owners of architectural-size plotters which emulate the

HP-Draftmaster can use the HP-Draftmaster option in the

plotter selection menu.

- Use 8.5"x11" paper on all plotters.

- All dot-matrix printer drivers were written for 80-

column printers.

- Graphics output to dot-matrix printers is not

presentation quality. This feature will be improved

upon in future releases of this software if there is

sufficient user demand.

- Owners of Epson compatible printers should use the

[Epson] option when selecting printers. Printers such

as the Panasonic KXP-1091 will work with this option.

- TI-855 printers should have the dip switch set in the

DATA-PROCESSING mode.

- Prior to printing or plotting a graph, check that the

correct device and output port have been set by

selecting SETTINGS from the CONFIGURE menu.

- When a serial port I/O error message is generated. It

is best to quit the program and restart.

The Data Analyst. ver 1.6 page 14

10. Tutorial

This section was written to guide a first-time user of the Data

Analyst through some of the more commonly used functions of the

program. The user is assumed to have a knowledge of DOS.

Begin by first copying DA.EXE and the accompanying sample data

files (with WDA extensions) to your hard disk, disk drive, or ram

drive. Be sure that you are logged to the drive and directory

containing the above files.

1. Run the program by typing DA

2. When the title screen is displayed, hit

to enter the FILE menu.

3. Use the up or down arrow keys to move the hi-lighted

area to LOAD. Hit the

4. You will be prompted for a file specification at the

bottom of the screen. Since you are already logged

into the proper directory, just hit the

5. A box containing all the file names in the current

directory will appear. Use the arrow keys to move the

hi-lighted cursor to the file DATA1.WDA (this is a

sample data file with two data sets). Hit the

key to load this file. Within a few moments the file

will be loaded and the FILE menu is again displayed.

6. Depress the right arrow key twice to select the GRAPH

menu. Use the up or down arrow keys to hi-light the

VIEW option and hit

mode. Pay close attention to the status boxes to the

right of the screen (described in the above section).

Notice that the number '2' appears in the NDS box and

the number '1' appears in the ADS box. This means that

two data sets have been loaded in memory and that the

current Active Data Set is data set #1. As can be seen

from the graph, data set #1 (symbol "o") contains ten

points and has been automatically scaled to fit the

entire graph area. A portion of data set #2 (symbol

"+") can be seen on the graph.

7. You can now familiarize yourself with some of the

functions available in this mode. Hit the F5 key once

to hide the grid. Hit F5 again to redisplay the grid.

Use F1 to enter the help screen. When you are back in

the graphics mode, use the arrow keys to move the

cursor located near the bottom of the graph. Note that

The Data Analyst. ver 1.6 page 15

the X and Y values on the display line correspond to

the current location of the cursor as if it were a data

point on the screen. Move the cursor to a position

near the center of the graph. Hit the F3 key to enter

the Label mode. The message box should display the

message "LABEL MODE". Type in the words "Sample Label"

and then hit the

labels are entered. Now hit F2 to preview the graph as

it would appear on hard copy. Depress any key to exit

from the preview screen. Back in the graphics mode,

hit the [C] key to erase the label you have just

entered (if you wish to do so).

8. To scale the graph to fit data set #2, first press the

[I] key. Notice the number in the ADS box has changed

from '1' to '2'. Press the [S] key and the graph will

automatically be rescaled to fit the second data set.

Note that you can repeat the procedure to scale the

graph to fit data set #1 again. Assuming that your

graph is still scaled to fit data set #2, hit the [F]

key to perform a linear regression fit on the Active

Data Set (which should be 2). The message 'Curve

fitting...' will appear briefly on the message box and

then disappear. A straight line will be drawn on the

graph. This is the least squares fit on data set #2.

To see the numerical results of the analysis, press F6.

Returning to the graphics mode, you will notice that

the regression line is gone. To redisplay the line,

press [P]. To perform the same analysis on data set

#1, press [I] to switch the ADS back to 1 (the ADS box

should show '1') and then press [F]. You can press F2

to preview the graph again from here.

9. As you may have already noticed, a linear fit on the

second data set is not very precise. A second order

polynomial fit will probably fit the data set better.

To do this, press

Hit the left arrow key once to enter the STATISTICS

menu. Press

second menu indicating the available regression types

will appear. Pick POLYNOMIAL and press

menu appears. Pick 2nd ORDER and press

by

menu. Press the right arrow key once to enter the

GRAPH menu. Pick VIEW and press

graphics mode a second time. Notice now near the bottom

right of the screen is the message Poly: 2. This means

that any regression analysis performed now will use the

2nd order polynomial fit.

The Data Analyst. ver 1.6 page 16

10. To perform a regression analysis on the second data set

again, first make sure that the ADS box is displaying

the number '2'. If not, press the [I] key. Now press

the [F] key to begin the analysis. A curved line

should appear within a few moments. This is the 2nd

order fit of data set #2. Press F2 to preview what you

have done so far. Note that you can have a different

type of regression analysis performed on each data set.

The resulting regression lines can be displayed or

removed. Press any key to return to the graphics mode.

11. To display the data points in the log scale, press F4.

Note that no autoscaling is possible in the Log mode.

The starting and ending Log cycle is user-defined in

the GRAPH menu in OPTIONS. Press F4 again to return to

the cartesian mode. For semi-log plots, press the [X]

or [Y] key. In all of these different scaling modes,

the regression lines that have be computed previously

for each data set can be redisplayed via the [P] key.

12. Press

arrow key once to get to the STATISTICS menu. Select

BASIC STATS and hit

Select DATA SET 1 and press

all of the basic statistical information on data set #1

will appear. Press

STATISTICS menu. You can repeat the above procedure to

obtain statistical data for the second data set. The

program will either issue a beep or a warning message

should the user attempt to obtain statistical

information on a nonexistent data set.

13. To exit the program, press

menu. Pick QUIT, hit

As you become more familiar with the program, you may decide to

customize more features in your graph by using the OPTIONS

selection in the GRAPH menu. You will also find that it is a

simple process to add or remove points in a data set by using the

built-in editor in the FILE menu.

*** Starting from version 1.2 of this program, the spacebar can be used

to select regression options in the graphics mode. Therefore, in

step 9 of this tutorial, the spacebar can be used to select a second

degree polynomial fit.

The Data Analyst. ver 1.6 page 17

11. User Feedback

It is only through user response that I can improve the program in

a direction that will benefit more users. All comments and

suggestions are welcomed. An EGA/VGA version of the program is

already in the works. Please direct all correspondence to me at

the address above or send E-mail to CompuServe (My CompuServe User

ID is 75136,2754).

12. General Facts

This program has been tested on the following machines:

IBM PC, IBM AT (6 and 8 Mhz), Zenith 386, TI BusinessPro,

Zenith 159, Zenith 148, AT&T 6300, IBM PS/2 30,

CompuAdd 286, CompuAdd 386, Gateway 2000 386-25MHz.

Most of the machines were fitted with the appropriate math

co-processor. Owners of 4.77Mhz machines with no co-processor

installed will probably find the program performance to be

inadequate.

Only users of DOS 3.0 or higher can use the SHELL command in the

Data Analyst without any risk. Users of DOS 2.0 or 2.1 will cause

the program to terminate unexpectedly when they invoke this option.

Generating regression fits on the Log or semi-Log scales may or

may not produce the regression line on-screen. This is due to the

fact that the resultant regression line may have negative x or y

values. In a similar vein, toggling back and forth between Log

and non-Log scales is not possible when negative data points

exist.

Users should be aware that the actual starting location of all

user-defined labels begin at the point of the small cursor when F3

is pressed (in the graphics mode). Use the X-Y coordinate display

of the cursor position as a guide to control the starting position

of the label more precisely.

The numerical and decimal point limitations of the editor

and scaling options have been noted. Work is being done on

a version of this program which circumvents these problems.

I wish to thank Anthony J. Shelton, Tze-Sen Lim, and Dr. James G.

Vaughan, all either presently or formerly with the Mechanical

Engineering Dept. at the University of Mississippi, for beta

testing the program as well as providing ideas for improving the

user interface. Ideas for additions to versions 1.2 - 1.5 are

attributed to Dr. Nadim Aziz of Clemson University.

The Data Analyst Registration/Order Form:

+---------------------------------------------------------------+

| |

| |

| To register, send this completed form along with |

| a contribution to: |

| |

| Kin-Peng Sin |

| P.O. Box 9316 |

| University, Mississippi 38677 |

| |

| Contributions of $20 or more entitles you to receive |

| the next release of this program free. |

| |

| |

| |

| |

| __________________________________________ |

| Name |

| |

| __________________________________________ |

| |

| __________________________________________ |

| Address |

| |

| _________________ __________ ________ |

| City State Zip |

| |

| |

| |

| Program version:_______________ (Displayed on title) |

| |

| |

| |

| Comments/Suggestions: |

| |

| ______________________________________________________ |

| |

| ______________________________________________________ |

| |

| ______________________________________________________ |

| |

| ______________________________________________________ |

| |

| ______________________________________________________ |

| |

| |

| |

+---------------------------------------------------------------+

Very nice! Thank you for this wonderful archive. I wonder why I found it only now. Long live the BBS file archives!

This is so awesome! 😀 I’d be cool if you could download an entire archive of this at once, though.

But one thing that puzzles me is the “mtswslnkmcjklsdlsbdmMICROSOFT” string. There is an article about it here. It is definitely worth a read: http://www.os2museum.com/wp/mtswslnk/