📄 m41 - producing descriptive statistics - 39 of 41.mht
字号:
<TD>Average</TD></TR>
<TR vAlign=3Dtop>
<TD noWrap> MIN</TD>
<TD>Minimum value</TD></TR>
<TR vAlign=3Dtop>
<TD noWrap> N</TD>
<TD>Number of observations with nonmissing values</TD></TR>
<TR vAlign=3Dtop>
<TD noWrap> NMISS</TD>
<TD>Number of observations with missing values</TD></TR>
<TR vAlign=3Dtop>
<TD noWrap> RANGE</TD>
<TD>Range</TD></TR>
<TR vAlign=3Dtop>
<TD noWrap> SKEWNESS</TD>
<TD>Skewness</TD></TR>
<TR vAlign=3Dtop>
<TD noWrap> STDDEV / STD </TD>
<TD>Standard Deviation</TD></TR>
<TR vAlign=3Dtop>
<TD noWrap> STDERR</TD>
<TD>Standard error of the mean</TD></TR>
<TR vAlign=3Dtop>
<TD noWrap> SUM</TD>
<TD>Sum</TD></TR>
<TR vAlign=3Dtop>
<TD noWrap> SUMWGT</TD>
<TD>Sum of the <CODE>Weight</CODE> variable values.</TD></TR>
<TR vAlign=3Dtop>
<TD noWrap> UCLM</TD>
<TD>One-sided confidence limit above the mean</TD></TR>
<TR vAlign=3Dtop>
<TD noWrap> USS</TD>
<TD>Uncorrected sum of squares</TD></TR>
<TR vAlign=3Dtop>
<TD noWrap> VAR</TD>
<TD>Variance</TD></TR></TBODY></TABLE>
<P><BR>Quantile Statistics=20
<TABLE title=3D"Quantile Statistics" cellSpacing=3D0 cellPadding=3D2 =
width=3D450=20
bgColor=3D#ffffff border=3D1>
<TBODY>
<TR bgColor=3D#f7f7de>
<TD><STRONG>Keyword</STRONG></TD>
<TD><STRONG>Description</STRONG></TD></TR>
<TR vAlign=3Dtop>
<TD noWrap> MEDIAN / P50 </TD>
<TD>Median or 50th percentile</TD></TR>
<TR vAlign=3Dtop>
<TD noWrap> P1</TD>
<TD>1st percentile</TD></TR>
<TR vAlign=3Dtop>
<TD noWrap> P5</TD>
<TD>5th percentile</TD></TR>
<TR vAlign=3Dtop>
<TD noWrap> P10</TD>
<TD>10th percentile</TD></TR>
<TR vAlign=3Dtop>
<TD noWrap> Q1 / P25</TD>
<TD>Lower quartile or 25th percentile</TD></TR>
<TR vAlign=3Dtop>
<TD noWrap> Q3 / P75</TD>
<TD>Upper quartile or 75th percentile</TD></TR>
<TR vAlign=3Dtop>
<TD noWrap> P90</TD>
<TD>90th percentile</TD></TR>
<TR vAlign=3Dtop>
<TD noWrap> P95</TD>
<TD>95th percentile</TD></TR>
<TR vAlign=3Dtop>
<TD noWrap> P99</TD>
<TD>99th percentile</TD></TR>
<TR vAlign=3Dtop>
<TD noWrap> QRANGE</TD>
<TD>Difference between upper and lower quartiles:=20
Q3-Q1</TD></TR></TBODY></TABLE>
<P><BR>Hypothesis Testing=20
<TABLE title=3D"Hypothesis Testing" cellSpacing=3D0 cellPadding=3D2 =
width=3D450=20
bgColor=3D#ffffff border=3D1>
<TBODY>
<TR bgColor=3D#f7f7de>
<TD><STRONG>Keyword</STRONG></TD>
<TD><STRONG>Description</STRONG></TD></TR>
<TR vAlign=3Dtop>
<TD noWrap> PROBT </TD>
<TD>Probability of a greater absolute value for the <EM>t</EM> =
value</TD></TR>
<TR vAlign=3Dtop>
<TD noWrap> T</TD>
<TD>Student's <EM>t</EM> for testing the hypothesis that the =
population=20
mean is 0</TD></TR></TBODY></TABLE></CENTER></CENTER>
<P align=3Dcenter><BR>
<CENTER>
<TABLE cellSpacing=3D0 cellPadding=3D0 width=3D"85%" border=3D0>
<TBODY>
<TR>
<TD><STRONG>Limiting Decimal Places</STRONG><BR>Because PROC MEANS =
uses=20
the BEST. format by default, procedure output can contain =
unnecessary=20
decimal places. To limit decimal places, use the <A=20
onmouseover=3D"window.status=3D'Display details on limiting =
decimal places.'; "=20
onclick=3DsuggestBack() onmouseout=3D"window.status=3D''"=20
=
href=3D"http://www.sas.com/apps/OLTRN/59080/m41/m41_7.htm">MAXDEC=3D=20
option</A> and set it equal to the length that you prefer.=20
<P><STRONG>Specifying Variables in PROC MEANS</STRONG><BR>By =
default, PROC=20
MEANS computes statistics for all numeric variables. To <A=20
onmouseover=3D"window.status=3D'Display details on specifying =
variables.'; "=20
onclick=3DsuggestBack() onmouseout=3D"window.status=3D''"=20
href=3D"http://www.sas.com/apps/OLTRN/59080/m41/m41_9.htm">specify =
the=20
variables</A> to include in PROC MEANS output, list them in a VAR=20
statement.=20
<P><STRONG>Group Processing Using the CLASS =
Statement</STRONG><BR>Include=20
a <A=20
onmouseover=3D"window.status=3D'Display details on the CLASS =
statement.'; "=20
onclick=3DsuggestBack() onmouseout=3D"window.status=3D''"=20
href=3D"http://www.sas.com/apps/OLTRN/59080/m41/m41_13.htm">CLASS=20
statement</A>, specifying variable names, to group PROC MEANS =
output by=20
variable values. Statistics are not computed for the CLASS =
variables.=20
<P><STRONG>Group Processing Using the BY=20
</STRONG><STRONG>Statement</STRONG><BR>Include a <A=20
onmouseover=3D"window.status=3D'Display details on the BY =
statement.'; "=20
onclick=3DsuggestBack() onmouseout=3D"window.status=3D''"=20
href=3D"http://www.sas.com/apps/OLTRN/59080/m41/m41_15.htm">BY=20
statement</A>, specifying variable names, to group PROC MEANS =
output by=20
variable values. Your data must be sorted according to those =
variables.=20
Statistics are not computed for the BY variables.=20
<P><STRONG>Creating a Summarized Data Set Using PROC =
MEANS</STRONG>=20
<BR>You can create an output data set that contains summarized =
variables=20
by using the <A=20
onmouseover=3D"window.status=3D'Display details on PROC MEANS.'; " =
onclick=3DsuggestBack() onmouseout=3D"window.status=3D''"=20
href=3D"http://www.sas.com/apps/OLTRN/59080/m41/m41_19.htm">OUTPUT =
statement=20
in PROC MEANS</A>. When you use the OUTPUT statement without =
specifying=20
the statistic-keyword=3D option, the summary statistics N, MEAN, =
STD, MIN,=20
and MAX are produced for all of the numeric variables or for all =
of the=20
variables that are listed in a VAR statement.=20
<P><STRONG>Creating a Summarized Data Set Using PROC =
SUMMARY</STRONG>=20
<BR>You can also create a summarized output data set by using <A=20
onmouseover=3D"window.status=3D'Display details on PROC SUMMARY.'; =
"=20
onclick=3DsuggestBack() onmouseout=3D"window.status=3D''"=20
href=3D"http://www.sas.com/apps/OLTRN/59080/m41/m41_20.htm">PROC=20
SUMMARY</A>. The PROC SUMMARY code for producing an output data =
set is=20
exactly the same as the code for producing an output data set with =
PROC=20
MEANS. The difference between the two procedures is that PROC =
MEANS=20
produces a report by default, whereas PROC SUMMARY produces an =
output data=20
set by default.=20
<P><STRONG>The FREQ Procedure</STRONG><BR>The <A=20
onmouseover=3D"window.status=3D'Display details on PROC SUMMARY.'; =
"=20
onclick=3DsuggestBack() onmouseout=3D"window.status=3D''"=20
href=3D"http://www.sas.com/apps/OLTRN/59080/m41/m41_23.htm">FREQ=20
Procedure</A> is a descriptive procedure as well as a statistical=20
procedure that produces one-way and n-way frequency tables. It =
concisely=20
describes your data by reporting the distribution of variable =
values.=20
<P><STRONG>Specifying Variables</STRONG><BR>By default, the FREQ =
procedure=20
creates frequency tables for every variable in your data set. To =
specify=20
the variables to analyze, include them in a <A=20
onmouseover=3D"window.status=3D'Display details on the TABLES =
statement.'; "=20
onclick=3DsuggestBack() onmouseout=3D"window.status=3D''"=20
=
href=3D"http://www.sas.com/apps/OLTRN/59080/m41/m41_25.htm">TABLES</A>=20
statement.=20
<P><STRONG>Creating Two-Way Tables<BR></STRONG>When a TABLES =
statement=20
contains two variables joined by an asterisk (*), PROC FREQ =
produces <A=20
onmouseover=3D"window.status=3D'Display details on =
crosstabulations.'; "=20
onclick=3DsuggestBack() onmouseout=3D"window.status=3D''"=20
=
href=3D"http://www.sas.com/apps/OLTRN/59080/m41/m41_29.htm">crosstabulati=
ons</A>.=20
The resulting table displays values for=20
<UL>
<LI>cell frequency=20
<LI>cell percentage of total frequency=20
<LI>cell percentage of row frequency=20
<LI>cell percentage of column frequency. </LI></UL>
<P><STRONG>Creating <EM>N</EM>-Way =
Tables<BR></STRONG>Crosstabulations can=20
include more than two variables. When <A=20
onmouseover=3D"window.status=3D'Display details on n-way tables.'; =
"=20
onclick=3DsuggestBack() onmouseout=3D"window.status=3D''"=20
href=3D"http://www.sas.com/apps/OLTRN/59080/m41/m41_32.htm">three =
or more=20
</A>variables are joined in a TABLES statement, the result is a =
series of=20
two-way tables that are grouped by the values of the first =
variables=20
listed.=20
<P><STRONG>Creating Tables in List Format</STRONG><BR>To reduce =
the bulk=20
of <EM>n</EM>-way table output, add a slash (/) and the LIST =
option to the=20
end of the TABLES statement. PROC FREQ then prints compact, <A=20
onmouseover=3D"window.status=3D'Display details on the LIST =
option.'; "=20
onclick=3DsuggestBack() onmouseout=3D"window.status=3D''"=20
=
href=3D"http://www.sas.com/apps/OLTRN/59080/m41/m41_34.htm">multi-column =
lists</A> instead of a series of tables.=20
<P><STRONG>Suppressing Table Information</STRONG><BR>You can =
suppress the=20
display of <A=20
onmouseover=3D"window.status=3D'Display details on suppressing =
statistics.'; "=20
onclick=3DsuggestBack() onmouseout=3D"window.status=3D''"=20
=
href=3D"http://www.sas.com/apps/OLTRN/59080/m41/m41_36.htm">specific=20
statistics</A> by adding one or more options to the TABLES =
statement:=20
<UL>
<LI><FONT color=3D#003399>NOFREQ</FONT> suppresses cell =
frequencies=20
<LI><FONT color=3D#003399>NOPERCENT</FONT> suppresses cell =
percentages=20
<LI><FONT color=3D#003399>NOROW</FONT> suppresses row =
percentages=20
<LI><FONT color=3D#003399>NOCOL</FONT> suppresses column =
percentages.=20
</LI></UL></TD></TR></TBODY></TABLE></CENTER><!-- =
InstanceEndEditable --><BR>
<TABLE cellSpacing=3D0 cellPadding=3D0 width=3D"85%" align=3Dcenter =
⌨️ 快捷键说明
复制代码
Ctrl + C
搜索代码
Ctrl + F
全屏模式
F11
切换主题
Ctrl + Shift + D
显示快捷键
?
增大字号
Ctrl + =
减小字号
Ctrl + -