@database Messlabor
@master Messlabor.english.guide
@$VER: Messlabor_V1.2_25.8.95
@author "Henk_Jonas"
@(c) "1995"
@smartwrap
@index index

@node "main" "CONTENTS"
@{CODE}
                        @{b}MESSLABOR@{ub}
                       ===========
 
                 Please make your choice:

                A short @{"Overview" link "kurz" 0}
                        @{"Requirements" link "req" 0}
                        @{"Copyrights" link "copy" 0}
                        @{"Installation" link "install" 0}

Explanation to the used @{"Fileformat" link "datei" 0}.
                A short @{"Example" link "bsp" 0}
        How to use this @{"Program" link "prog" 0}?
                        @{"Preferences" link "prefs" 0}

                        @{"Credits" link "credits" 0}
                        @{"Author" link "adresse" 0}
                        @{"Tables" link "statistik" 0}
                        @{"History" link "history" 0}
@{BODY}

Please excuse my poor english.

@endnode

@node "kurz" "short overview"
This program allowed to make plots from given x/y datas. You can link the
points with different modes (nothing, line, spline) and apply some @{"statistic" link "statistik" 0}
functions (like regression, median, mean etc.) on the datas.

@endnode

@node "req" "Requirements"
@{CODE}
  - WB 3.0 or better (should not be any problem)
  - at least 1 MByte RAM (more recommended!)
  - iff.library 2.0
@{BODY}

@endnode

@node "copy" "Copyrights:"
Messlabor is Copyright by Henk Jonas. The complete archiv can be freely
distributed. I give you the permission to change this documentation and the
catalog-files in case you found an error, a bad explanation or you want
translate it in another language. But, please send me your changes.

The distributor may charge a fee to recover distribution costs. But not more!

I make no warranties or representations with this software. I hope there are
no errors but i don't guarantee it. If you have found an error or you have
any idea what this program should can do, please write me a letter or EMail.
(@{"address" link "adresse" 0})

I'm very happy if you send gifts, cards or money to me, or if you support
independent and progressive groups.

@endnode

@node "install" "Installation"
Just copy the whole drawer wherever you want.
Change in the programicon the @{"tooltypes" LINK "prefs" 37} for @{b}GUIDE=@{ub} to
@{b}GUIDE=PROGDIR:Messlabor.english.guide@{ub} and for @{b}SCREEN=@{ub} to
@{b}SCREEN=0@{ub}. After the program is started you can use the menu
@{b}Others/Preferences...@{ub} to change all interesting @{"settings" LINK "prefs" 0}.

@endnode

@node "datei" "Fileformat:"
The program support ASCII-files. I think about support for other fileformats,
if you have an idea for a wellsupported fileformat please write me. (@{"address" link "adresse" 0})

The datas can be in column- or lineformat. The x- and y-datas can change in
the programm. The titles of axes will take from the first cells in the
table, but you can edit this titles w/o to change the contents in the first
cells.

Numbers is'nt allowed to open with blanks. The separators for columns and
lines and the decimalpoint may free change.

(@{"Example" link "bsp" 0})

@endnode

@node "bsp" "Examplefile:"
Here you can see a example:


@{CODE}
time,way
0,0
1,3
2,7
3,9
4,12
5,11
6,6
7,4
8,0
@{BODY}


This file can loaded with this @{"Preferences:" link "prefs" 0}


@{CODE}
ZEILENTRENNER=10      (lineseparator)
SPALTENTRENNER=44     (columnseparator)
DEZIMALPUNKT=46       (decimalpoint)
SPALTEN=2             (columnes)
ZEILEN=10             (lines)
@{BODY}


You find in this archiv also two other examples, this can also be loaded with
the (default) preferences above. Please look also at the plots.


@{CODE}
· @{"Example 1" link "Beispiel1.dat/main" 0}
· @{"Example 2" link "Beispiel2.dat/main" 0}
· @{"Plot to example 1" link "Beispiel1.iff/main" 0}
· @{"Plot to example 2" link "Beispiel2.iff/main" 0}
@{BODY}

@endnode

@node "prog" "How to use?"
Let's split this program in two sections. First the @{"Tablesection" link "eins" 0},
here you can see the datas in a table and you can select which columns you
want to use, and second the @{"Plotsection" link "zwei" 0}, here you can plot
the datas.

@endnode

@node "eins" "Tablesection:"
The menus:


@{b}Project@{ub}

- @{b}New@{ub} Clear the table.

- @{b}Open-->ASCII@{ub} Load a ASCII-(Text) @{"file" link "datei" 0} with
the datas.

- @{b}Save as...@{ub} Save a ASCII-(Text) @{"file" link "datei" 0} under a
giving name.

- @{b}About@{ub} Infos about version and @{"Programmer" link "adresse" 0}.

- @{b}Quit@{ub} Quit the program.


@{b}Clipboard@{ub}

- @{b}Copy@{ub} Copy the table with columnseparator (TABULATOR) and
lineseparator (RETURN) into the Clipboard.

- @{b}Paste@{ub} Paste datas from the clipboard into the table.


@{b}Edit@{ub}

- @{b}Transpose@{ub} Change columns with lines in the table.

- @{b}x-datas@{ub} Select column with x-datas.

- @{b}1.y-datas@{ub} Select column with main-y-datas for use in the plots
and for use with the @{"statistic-functions" link "zwei" 0}.

- @{b}2.y-datas@{ub} Select column with second-y-datas.

- @{b}Plot@{ub} Plot after query some @{"settings" link "diawin" 0} the x/y
plot.


@{"statistic" link "statistik" 0}

- @{b}Parameter-->Sort Y to Y2@{ub} Copy the column mark with Y sorted in
the column mark with Y2. The column Y2 will be cleared before.

- @{b}Parameter-->Sort for X@{ub} Sort all columns in the table for the
column mark with X.

- @{b}Parameter-->Parameter from X@{ub} Calculate statistic parameters from
the column mark with X. You must sort the columne before to get the right
parameters!

- @{b}Analysis-->T-Test@{ub} Please look at @{"statistic" link "statistik" 0}
from X/Y means that you test the columns X and Y, from para. means that you
can insert the means, std. dev. and number of observation for both datas.

- @{b}Analysis-->Analysis of var.@{ub} Please look at @{"statistic" link "statistik" 0}
X-Y means that you analise the columns between X and Y, para. are n.y.i.

- @{b}Analysis-->Chi-Square-Test@{ub} not yet implemented

- @{b}Links-->Correlation@{ub} not yet implemented

- @{b}Links-->Regression@{ub} Calculate @{"Regression" link "statistik" 0}
as Y-datas= function from (X-datas). With function yet a polynom of
1 to 5 degree.


@{b}Others@{ub}

- @{b}Preferences...@{ub} Open the @{"Preference" link "prefs" 0}-window,
for screenmode, column- and lineseparator, decimalpoint and tabledimension.

- @{b}Help@{ub} Open this helpfile.


The Gadgets 'Open', 'X-->', 'Y-->', 'Y2->', 'Plot' stand for the Menuitems
@{b}Open-->ASCII@{ub}, @{b}Edit/x-datas@{ub}, @{b}Edit/1.y-datas@{ub},
@{b}Edit/2.y-datas@{ub} und @{b}Edit/Plot@{ub}.

With the cursorkeys or the mouse you can mark the @{b}active cell@{ub} in the
table and the @{b}active column@{ub}. If you press any key, you activate
the stringgadget over the table, now it's possible to change the contents of
the active cell.

@endnode

@node "diawin" "Plotsettings:"
window: @{b}Preference for Plot@{ub}


  This @{"Window" link "fenster1.e/main" 0} will be opened if you press the
@{b}Plot@{ub} Gadget or use the menuitem @{b}Edit/Plot@{ub}.


here are some gadgets:

1. row: the titletext for the plot and the font for the titletext

2. row: the first line of the x-axistext and the y-axistext and a button
for "exist x-datas for the y-datas?" n.i.y.

3. row: the second line of the x-axistext and the y-axistext and a button
for "show datatable on the plot?" n.i.y.

4. row: factors for the x- and y-datas and a button for "show zeropoint?"

5. row: a button for "regressionplot on?" and a cyclegadget for "how to
link the points?"

6. row: a cyclegadget for "which color for the plot?" n.i.y.

7. row: the font for the axistexts

8. row: buttons for "OK" and "CANCEL"

@endnode

@node "zwei" "Plotsection:"
The menus:


@{b}Project@{ub}

- @{b}New@{ub}   Force a newdraw of the plot

- @{b}Save-->as ILBM...@{ub}   Save the plot as ILBM(IFF) with free width and
height

- @{b}Save-->as GEM...@{ub}    Save the plot as GEM-Metafile

- @{b}Save-->as CGM...@{ub}    Save the plot as CGM-Metafile

- @{b}Save-->as EPS...@{ub}    Save the plot as EPS

- @{b}Print@{ub}   Print the plot with full printerresolution

- @{b}About @{ub}

- @{b}Close@{ub}   Close the plot


@{b}Clipboard@{ub}

- @{b}Copy@{ub}    Copy the plot in the actually size to Clipboard, Unit 0


@{b}Statistic@{ub}

- @{b}Regression@{ub}  Calculate a regressionpolynom with degree from 1 to 5

After the calculation you can change the parameter and see the error for the
regressionpolynom.


@{b}Misc@{ub}

- @{b}Preferences...@{ub}  Open the window: @{"Preference for plot" link "diawin" 0}

- @{b}Help@{ub}  Open this helpfile

@endnode

@node "prefs" "Preferences:"
@{b}Preferences@{ub}


  All preferences can change in the program or in the tooltypes in the
icon.


@{b}max. tablesize@{ub}

  Here you can change the size of the table. The tablesize are static, that
means the program will allocated the whole memory for the table to the begin.
You need 20*lines*columns [Byte]


@{b}separators@{ub}

  Here you can set the separators that the program use if it load a
@{"ASCII-file" link "datei" 0}. The decimalpoint will be use in the whole
plot as decimalpoint.

  Behind the gadgets to insert the ASCII-CODES, exist a place to show the
sign that you set. If you see a square, it means, the sign are not visible
(z.B. newline or tabulator).


@{b}misc@{ub}

Here you can change the screenmode and the place where are this helpfile.


To use the changes of screenmode or tablesize, you must restart the program.


The changes will be saved into the icon. You can change the tooltypes also
separate.


There are this tooltypes possible:

@{CODE}
SCREEN=           (screenmode)
ZEILENTRENNER=    (ASCII-Code for the lineseparator)
SPALTENTRENNER=   (ASCII-Code for the columnseparator)
DEZIMALPUNKT=     (ASCII-Code for the decimalpoint)
SPALTEN=          (max. columnsize)
ZEILEN=           (max. linesize)
GUIDE=            (name and path for this helpfile)
@{BODY}

  Please use for SCREEN the number of the @{"screenmode" link "modis" 0}
or 0 for to use the workbench.


That's the defaults:


@{CODE}
ZEILENTRENNER=10  (normaly RETURN)
SPALTENTRENNER=44 (,)
DEZIMALPUNKT=46   (.)
SPALTEN=10
ZEILEN=100
SCREEN=0          (Workbench Screen)
GUIDE=PROGDIR:Meßlabor.guide
@{BODY}

@endnode

@node "modis" "Examples for screenmodes:"
@{CODE}
PAL:Hires Lace =$29004
NTSC:Hires Lace=$19004
MULTISCAN:Prod.=$39024
@{BODY}

@endnode

@node "Statistik" "Statistic"
parameters:

@{CODE}
@{b}Kennwerte:@{ub} (please sort the datas before)

@{b}mean@{ub}
@{b}median@{ub}
@{b}quantile@{ub}
@{b}number of observations@{ub}
@{b}standard dev@{ub}
@{b}error of mean@{ub}
@{b}gausstest@{ub}
@{BODY}


tests:

@{CODE}
- @{b}t-test@{ub}
- @{b}analyse of variance@{ub}
@{BODY}


tables:

@{CODE}
- table for @{"gausstest" link "TAB1" 0}
- table for @{"t-test" link "TAB2" 0}
- table for @{"variances" link "TAB3" 0}
@{BODY}

@endnode

@node "TAB1" "gausstest"
the parameter must be between:


@{CODE}
(for p=0.1)

  n           >           <

  5          2.22        2.71
  6          2.37        2.94
  7          2.49        3.14
  8          2.59        3.30
  9          2.68        3.45

 10          2.76        3.57
 11          2.84        3.68
 12          2.90        3.78
 13          2.96        3.87
 14          3.02        3.95
 15          3.07        4.02
 16          3.12        4.09
 17          3.17        4.15
 18          3.21        4.21
 19          3.25        4.27

 20          3.29        4.32
 25          3.45        4.53
 30          3.59        4.70
 40          3.79        4.96
 50          3.95        5.14
 60          4.08        5.29
 70          4.19        5.41
100          4.44        5.68
150          4.72        5.96
200          4.90        6.15
@{BODY}

@endnode

@node "TAB2" "t-test"
if parameter lower then tablevalue, then you have the same observations


@{CODE}
degree       oneside            twoside
  of
freedom  p=0.05    p=0.01    p=0.05 p=0.01

  1       6.31     31.82    12.71     63.66
  2       2.31      6.97     4.30      9.92
  3       2.35      4.54     3.18      5.84
  4       2.13      3.75     2.78      4.60
  5       2.02      3.37     2.57      4.03
  6       1.94      3.14     2.45      3.71
  7       1.89      3.00     2.36      3.50
  8       1.86      2.90     2.31      3.36
  9       1.83      2.82     2.26      3.25
 10       1.81      2.76     2.23      3.17
 15       1.75      2.60     2.13      2.95
 20       1.72      2.53     2.09      2.85
 30       1.70      2.46     2.04      2.75
 50       1.68      2.40     2.01      2.68
100       1.66      2.37     1.98      2.63
@{BODY}

@endnode

@node "TAB3" "f-ratio"
if parameter lower then tablevalue, then you have the same observations

if you look for variance-test (inside the t-test) then mean a parameter
lower then the tablevalue pooled variance

'analysisof variance: x with (n1,n2) degrees of freedom'


@{CODE}
for p=0.05

    n1  1    2    3    4    5    6    7    8    9   10    15   20   30   50 100
 n2
  1    161  200  216  225  230  234  237  239  241  242  246  248  250  252 253
  2   18.5 19.0 19.2 19.2 19.3 19.3 19.4 19.4 19.4 19.4 19.4 19.4 19.5 19.5 19.5
  3   10.1 9.55 9.28 9.12 9.01 8.94 8.89 8.85 8.81 8.79 8.70 8.66 8.62 8.58 8.55
  4   7.71 6.94 6.59 6.39 6.26 6.16 6.09 6.04 6.00 5.96 5.86 5.80 5.75 5.70 5.66
  5   6.61 5.79 5.41 5.19 5.05 4.95 4.88 4.82 4.77 4.74 4.62 4.56 4.50 4.44 4.41
  7   5.99 4.74 4.35 4.12 3.97 3.87 3.79 3.73 3.68 3.64 3.51 3.44 3.38 3.32 3.27
 10   4.96 4.10 3.71 3.48 3.33 3.22 3.14 3.07 3.02 2.98 2.85 2.77 2.70 2.64 2.59
 15   4.54 3.68 3.29 3.06 2.90 2.79 2.71 2.64 2.59 2.54 2.40 2.33 2.25 2.18 2.12
 20   4.35 3.49 3.10 2.87 2.71 2.60 2.51 2.45 2.39 2.35 2.20 2.12 2.04 1.97 1.70
 30   4.17 3.32 2.92 2.69 2.53 2.42 2.33 2.27 2.21 2.16 2.01 1.93 1.84 1.76 1.70
 50   4.03 3.18 2.79 2.56 2.40 2.29 2.20 2.13 2.07 2.03 1.87 1.78 1.69 1.60 1.52
100   3.94 3.09 2.70 2.46 2.31 2.19 2.10 2.03 1.97 1.93 1.77 1.68 1.57 1.48 1.39
@{BODY}

@endnode

@node "TAB4" "Tabelle 4"

@endnode

@node "credits" "Danksagungen:"
 Ich bedanke mich bei:


Stefan Le Breton für das Amiga RKRM Devices, den Beispielcode für EPS-
Ausgabe und Betatesting?

Piet Jonas für die Idee (er hat ein ähnliches Programm für WinDoof geschrieben)
und den Tip bei der Regression.

Heiko Bergmann für seine Suche im WWW nach Infos über das CGM-Format.

Dietmar Eilert für seinen tollen Editor.

Wouter van Oortmerssen für seinen E-Compiler (wenn mir ganz viel Leute
ganz viel Geld für dieses Programm schicken, werde ich mir vielleicht ja auch
mal eine neuere Version, als die 2.1b zulegen können)

Haike für ihre riesige Geduld (das waren viele Stunden am Rechner)


Desweiteren einen herzlichen Dank an:

 Swen Stullich (SWEN@trashcan.mcnet.de) und

 Lars Bischoff (Lars_Bischoff@hb2.maus.de),

die sich mit Verbesserungsvorschlägen und Kommentaren bei mir meldeten.

sowie an

 Jürgen Brosi (Nero@highvolt.gun.de),

für seine Vorschläge und die 20,-DM

@endnode

@node "adresse" "Adresse des Autors:"
@{CODE}
SnailMail: Henk Jonas, Zionskirchstr. 28, 10119 Berlin, Germany
EMail: subvcbhd@w271zrz.zrz.tu-berlin.de
@{BODY}

@endnode

@node "history" "Programmentwicklung"
 folgende Fehler/Funktionen wurden beseitigt/eingeführt



@{CODE}
V1.0     erste Veröffentlichung

V1.0b    - Fehler bei StringToReal beseitigt
         - Regression zeigt Koeffizienten an
         - Regressionskoeffizienten können verändert werden
         - AmigaGuideLibrary wurde zweimal geschlossen (:-O)

V1.1     - PubScreen "Worbench" wieder freigeben, wenn nicht mehr gebraucht
         - mehr Statistik-Fkt. (Mittel, Median, ...)
         - Regressionskurve malt nicht mehr außerhalb des Diagramms

V1.1b    - Werte im Programm änderbar
         - EPS-Ausgabe des Diagramms
         - Parameterübergabe durch die Workbench
         - Position des Fensters wird gespeichert

V1.2     - EPS-Ausgabe bis auf Vorschaubild vollständig
         - Localisiert (bisher gibt's einen englischen Katalog)
@{BODY}

@endnode

@node "INDEX" "index"
@{CODE}

@{"Overview" link "kurz" 0}
@{"Requirements" link "req" 0}
@{"Copyrights" link "copy" 0}
@{"Fileformat" link "datei" 0}
@{"Examplefile" link "bsp" 0}
@{"How to use it?" link "prog" 0}
@{"Tablesection" link "eins" 0}
@{"Preferences for plot" link "diawin" 0}
@{"Plotsection" link "zwei" 0}
@{"Statistic" link "Statistik" 0}
@{"Preferences" link "prefs" 0}
@{"Tooltypes" link "prefs" 0}
@{"Examples for screenmodes" link "modis" 0}
@{"Credits" link "credits" 0}
@{"my address" link "adresse" 0}
@{"History" link "history" 0}
@{BODY}

@endnode


