freebsd-dev/usr.bin/ministat/ministat.1
Eitan Adler aa374634b4 Add option to suppress just the plot in ministat while still retaining
the relative comparison (i.e., useful part).

Approved by:	cperciva
MFC after:	3 days
2012-11-15 15:06:12 +00:00

134 lines
4.9 KiB
Groff

.\"
.\" Copyright (c) 2007 Poul-Henning Kamp
.\" All rights reserved.
.\"
.\" Redistribution and use in source and binary forms, with or without
.\" modification, are permitted provided that the following conditions
.\" are met:
.\" 1. Redistributions of source code must retain the above copyright
.\" notice, this list of conditions and the following disclaimer.
.\" 2. Redistributions in binary form must reproduce the above copyright
.\" notice, this list of conditions and the following disclaimer in the
.\" documentation and/or other materials provided with the distribution.
.\"
.\" THIS SOFTWARE IS PROVIDED BY THE AUTHOR AND CONTRIBUTORS ``AS IS'' AND
.\" ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
.\" IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE
.\" ARE DISCLAIMED. IN NO EVENT SHALL THE AUTHOR OR CONTRIBUTORS BE LIABLE
.\" FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL
.\" DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS
.\" OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION)
.\" HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT
.\" LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY
.\" OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF
.\" SUCH DAMAGE.
.\"
.\" $FreeBSD$
.\"
.Dd November 10, 2012
.Dt MINISTAT 1
.Os
.Sh NAME
.Nm ministat
.Nd statistics utility
.Sh SYNOPSIS
.Nm
.Op Fl Ans
.Op Fl C Ar column
.Op Fl c Ar confidence_level
.Op Fl d Ar delimiter
.Op Fl w Op width
.Op Ar
.Sh DESCRIPTION
The
.Nm
command calculates fundamental statistical properties of numeric data
in the specified files or, if no file is specified, standard input.
.Pp
The options are as follows:
.Bl -tag -width Fl
.It Fl A
Just report the statistics of the input and relative comparisons,
suppress the ASCII-art plot.
.It Fl n
Just report the raw statistics of the input, suppress the ASCII-art plot
and the relative comparisons.
.It Fl s
Print the average/median/stddev bars on separate lines in the ASCII-art
plot, to avoid overlap.
.It Fl C Ar column
Specify which column of data to use.
By default the first column in the input file(s) are used.
.It Fl c Ar confidence_level
Specify desired confidence level for Student's T analysis.
Possible values are 80, 90, 95, 98, 99 and 99.5 %
.It Fl d Ar delimiter
Specifies the column delimiter characters, default is SPACE and TAB.
See
.Xr strtok 3
for details.
.It Fl w Ar width
Width of ASCII-art plot in characters, default is 74.
.El
.Pp
A sample output could look like this:
.Bd -literal -offset indent
$ ministat -s -w 60 iguana chameleon
x iguana
+ chameleon
+------------------------------------------------------------+
|x * x * + + x +|
| |________M______A_______________| |
| |________________M__A___________________| |
+------------------------------------------------------------+
N Min Max Median Avg Stddev
x 7 50 750 200 300 238.04761
+ 5 150 930 500 540 299.08193
No difference proven at 95.0% confidence
.Ed
.Pp
If
.Nm
tells you, as in the example above, that there is no difference
proven at 95% confidence, the two data sets you gave it are for
all statistical purposes identical.
.Pp
You have the option of lowering your standards by specifying a
lower confidence level:
.Bd -literal -offset indent
$ ministat -s -w 60 -c 80 iguana chameleon
x iguana
+ chameleon
+------------------------------------------------------------+
|x * x * + + x +|
| |________M______A_______________| |
| |________________M__A___________________| |
+------------------------------------------------------------+
N Min Max Median Avg Stddev
x 7 50 750 200 300 238.04761
+ 5 150 930 500 540 299.08193
Difference at 80.0% confidence
240 +/- 212.215
80% +/- 70.7384%
(Student's t, pooled s = 264.159)
.Ed
.Pp
But a lower standard does not make your data any better, and the
example is only included here to show the format of the output when
a statistical difference is proven according to Student's T method.
.Sh SEE ALSO
Any mathematics text on basic statistics, for instances Larry Gonicks
excellent "Cartoon Guide to Statistics" which supplied the above example.
.Sh HISTORY
The
.Nm
command was written by Poul-Henning Kamp out of frustration
over all the bogus benchmark claims made by people with no
understanding of the importance of uncertainty and statistics.
.Pp
From
.Fx 5.2
it has lived in the source tree as a developer tool, graduating
to the installed system from
.Fx 8.0 .