18f2c52e6SPoul-Henning Kamp.\" 28f2c52e6SPoul-Henning Kamp.\" Copyright (c) 2007 Poul-Henning Kamp 38f2c52e6SPoul-Henning Kamp.\" All rights reserved. 48f2c52e6SPoul-Henning Kamp.\" 58f2c52e6SPoul-Henning Kamp.\" Redistribution and use in source and binary forms, with or without 68f2c52e6SPoul-Henning Kamp.\" modification, are permitted provided that the following conditions 78f2c52e6SPoul-Henning Kamp.\" are met: 88f2c52e6SPoul-Henning Kamp.\" 1. Redistributions of source code must retain the above copyright 98f2c52e6SPoul-Henning Kamp.\" notice, this list of conditions and the following disclaimer. 108f2c52e6SPoul-Henning Kamp.\" 2. Redistributions in binary form must reproduce the above copyright 118f2c52e6SPoul-Henning Kamp.\" notice, this list of conditions and the following disclaimer in the 128f2c52e6SPoul-Henning Kamp.\" documentation and/or other materials provided with the distribution. 138f2c52e6SPoul-Henning Kamp.\" 148f2c52e6SPoul-Henning Kamp.\" THIS SOFTWARE IS PROVIDED BY THE AUTHOR AND CONTRIBUTORS ``AS IS'' AND 158f2c52e6SPoul-Henning Kamp.\" ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE 168f2c52e6SPoul-Henning Kamp.\" IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE 178f2c52e6SPoul-Henning Kamp.\" ARE DISCLAIMED. IN NO EVENT SHALL THE AUTHOR OR CONTRIBUTORS BE LIABLE 188f2c52e6SPoul-Henning Kamp.\" FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL 198f2c52e6SPoul-Henning Kamp.\" DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS 208f2c52e6SPoul-Henning Kamp.\" OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) 218f2c52e6SPoul-Henning Kamp.\" HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT 228f2c52e6SPoul-Henning Kamp.\" LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY 238f2c52e6SPoul-Henning Kamp.\" OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF 248f2c52e6SPoul-Henning Kamp.\" SUCH DAMAGE. 258f2c52e6SPoul-Henning Kamp.\" 268f2c52e6SPoul-Henning Kamp.\" $FreeBSD$ 278f2c52e6SPoul-Henning Kamp.\" 288f2c52e6SPoul-Henning Kamp.Dd December 20, 2007 298f2c52e6SPoul-Henning Kamp.Dt MINISTAT 1 308f2c52e6SPoul-Henning Kamp.Os 318f2c52e6SPoul-Henning Kamp.Sh NAME 328f2c52e6SPoul-Henning Kamp.Nm ministat 338f2c52e6SPoul-Henning Kamp.Nd statistics utility 348f2c52e6SPoul-Henning Kamp.Sh SYNOPSIS 358f2c52e6SPoul-Henning Kamp.Nm 368f2c52e6SPoul-Henning Kamp.Op Fl ns 378f2c52e6SPoul-Henning Kamp.Op Fl C Ar column 388f2c52e6SPoul-Henning Kamp.Op Fl c Ar confidence_level 398f2c52e6SPoul-Henning Kamp.Op Fl d Ar delimiter 408f2c52e6SPoul-Henning Kamp.Op Fl w Op width 418f2c52e6SPoul-Henning Kamp.Op Ar 428f2c52e6SPoul-Henning Kamp.Sh DESCRIPTION 438f2c52e6SPoul-Henning KampThe 448f2c52e6SPoul-Henning Kamp.Nm 458f2c52e6SPoul-Henning Kampcommand calculates fundamental statistical properties of numeric data 468f2c52e6SPoul-Henning Kampin the specified files or, if no file is specified, standard input. 478f2c52e6SPoul-Henning Kamp.Pp 488f2c52e6SPoul-Henning KampThe options are as follows: 498f2c52e6SPoul-Henning Kamp.Bl -tag -width Fl 508f2c52e6SPoul-Henning Kamp.It Fl n 518f2c52e6SPoul-Henning KampJust report the raw statistics of the input, suppress the ASCII-art plot 528f2c52e6SPoul-Henning Kampand the relative comparisons. 538f2c52e6SPoul-Henning Kamp.It Fl s 548f2c52e6SPoul-Henning KampPrint the average/median/stddev bars on separate lines in the ASCII-art 558f2c52e6SPoul-Henning Kampplot, to avoid overlap. 568f2c52e6SPoul-Henning Kamp.It Fl C Ar column 578f2c52e6SPoul-Henning KampSpecify which column of data to use. 588f2c52e6SPoul-Henning KampBy default the first column in the input file(s) are used. 598f2c52e6SPoul-Henning Kamp.It Fl c Ar confidence_level 608f2c52e6SPoul-Henning KampSpecify desired confidence level for Student's T analysis. 61d60522ebSPeter WemmPossible values are 80, 90, 95, 98, 99 and 99.5 % 628f2c52e6SPoul-Henning Kamp.It Fl d Ar delimiter 638f2c52e6SPoul-Henning KampSpecifies the column delimiter characters, default is SPACE and TAB. 648f2c52e6SPoul-Henning KampSee 658f2c52e6SPoul-Henning Kamp.Xr strtok 3 668f2c52e6SPoul-Henning Kampfor details. 678f2c52e6SPoul-Henning Kamp.It Fl w Ar width 688f2c52e6SPoul-Henning KampWidth of ASCII-art plot in characters, default is 74. 698f2c52e6SPoul-Henning Kamp.El 708f2c52e6SPoul-Henning Kamp.Pp 718f2c52e6SPoul-Henning KampA sample output could look like this: 728f2c52e6SPoul-Henning Kamp.Bd -literal -offset indent 738f2c52e6SPoul-Henning Kamp $ ministat -s -w 60 iguana chameleon 748f2c52e6SPoul-Henning Kamp x iguana 758f2c52e6SPoul-Henning Kamp + chameleon 768f2c52e6SPoul-Henning Kamp +------------------------------------------------------------+ 778f2c52e6SPoul-Henning Kamp |x * x * + + x +| 788f2c52e6SPoul-Henning Kamp | |________M______A_______________| | 798f2c52e6SPoul-Henning Kamp | |________________M__A___________________| | 808f2c52e6SPoul-Henning Kamp +------------------------------------------------------------+ 818f2c52e6SPoul-Henning Kamp N Min Max Median Avg Stddev 828f2c52e6SPoul-Henning Kamp x 7 50 750 200 300 238.04761 838f2c52e6SPoul-Henning Kamp + 5 150 930 500 540 299.08193 848f2c52e6SPoul-Henning Kamp No difference proven at 95.0% confidence 858f2c52e6SPoul-Henning Kamp.Ed 868f2c52e6SPoul-Henning Kamp.Pp 878f2c52e6SPoul-Henning KampIf 888f2c52e6SPoul-Henning Kamp.Nm 898f2c52e6SPoul-Henning Kamptells you, as in the example above, that there is no difference 908f2c52e6SPoul-Henning Kampproven at 95% confidence, the two data sets you gave it are for 918f2c52e6SPoul-Henning Kampall statistical purposes identical. 928f2c52e6SPoul-Henning Kamp.Pp 938f2c52e6SPoul-Henning KampYou have the option of lowering your standards by specifying a 948f2c52e6SPoul-Henning Kamplower confidence level: 958f2c52e6SPoul-Henning Kamp.Bd -literal -offset indent 968f2c52e6SPoul-Henning Kamp $ ministat -c 80 iguana chameleon 978f2c52e6SPoul-Henning Kamp x iguana 988f2c52e6SPoul-Henning Kamp + chameleon 998f2c52e6SPoul-Henning Kamp +------------------------------------------------------------+ 1008f2c52e6SPoul-Henning Kamp |x * x * + + x +| 1018f2c52e6SPoul-Henning Kamp | |________M______A_______________| | 1028f2c52e6SPoul-Henning Kamp | |________________M__A___________________| | 1038f2c52e6SPoul-Henning Kamp +------------------------------------------------------------+ 1048f2c52e6SPoul-Henning Kamp N Min Max Median Avg Stddev 1058f2c52e6SPoul-Henning Kamp x 7 50 750 200 300 238.04761 1068f2c52e6SPoul-Henning Kamp + 5 150 930 500 540 299.08193 1078f2c52e6SPoul-Henning Kamp Difference at 80.0% confidence 1088f2c52e6SPoul-Henning Kamp 240 +/- 212.215 1098f2c52e6SPoul-Henning Kamp 80% +/- 70.7384% 1108f2c52e6SPoul-Henning Kamp (Student's t, pooled s = 264.159) 1118f2c52e6SPoul-Henning Kamp.Ed 1128f2c52e6SPoul-Henning Kamp.Pp 1138f2c52e6SPoul-Henning KampBut a lower standard does not make your data any better, and the 1148f2c52e6SPoul-Henning Kampexample is only included here to show the format of the output when 1158f2c52e6SPoul-Henning Kampa statistical difference is proven according to Student's T method. 1168f2c52e6SPoul-Henning Kamp.Sh SEE ALSO 1178f2c52e6SPoul-Henning KampAny mathematics text on basic statistics, for instances Larry Gonicks 1188f2c52e6SPoul-Henning Kampexcellent "Cartoon Guide to Statistics" which supplied the above example. 1198f2c52e6SPoul-Henning Kamp.Sh HISTORY 1208f2c52e6SPoul-Henning KampThe 1218f2c52e6SPoul-Henning Kamp.Nm 1228f2c52e6SPoul-Henning Kampcommand was written by Poul-Henning Kamp out of frustration 1238f2c52e6SPoul-Henning Kampover all the bogus benchmark claims made by people with no 1248f2c52e6SPoul-Henning Kampunderstanding of the importance of uncertainty and statistics. 1258f2c52e6SPoul-Henning Kamp.Pp 1268f2c52e6SPoul-Henning KampFrom 1278f2c52e6SPoul-Henning Kamp.Fx 5.2 1288f2c52e6SPoul-Henning Kampit has lived in the source tree as a developer tool, graduating 1298f2c52e6SPoul-Henning Kampto the installed system from 1308f2c52e6SPoul-Henning Kamp.Fx 8.0 . 131