1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
|
<?xml version="1.0" encoding="UTF-8"?>
<helpdocument version="1.0">
<!--
***********************************************************************
*
* DO NOT ALTER OR REMOVE COPYRIGHT NOTICES OR THIS FILE HEADER.
*
* Copyright 2000, 2010 Oracle and/or its affiliates.
*
* OpenOffice.org - a multi-platform office productivity suite
*
* This file is part of OpenOffice.org.
*
* OpenOffice.org is free software: you can redistribute it and/or modify
* it under the terms of the GNU Lesser General Public License version 3
* only, as published by the Free Software Foundation.
*
* OpenOffice.org is distributed in the hope that it will be useful,
* but WITHOUT ANY WARRANTY; without even the implied warranty of
* MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
* GNU Lesser General Public License version 3 for more details
* (a copy is included in the LICENSE file that accompanied this code).
*
* You should have received a copy of the GNU Lesser General Public License
* version 3 along with OpenOffice.org. If not, see
* <http://www.openoffice.org/license.html>
* for a copy of the LGPLv3 License.
*
************************************************************************
-->
<meta>
<topic id="textschart0104050100xml" indexer="include" status="PUBLISH">
<title xml-lang="en-US" id="tit">Trend Lines </title>
<filename>/text/schart/01/04050100.xhp</filename>
</topic>
</meta>
<body>
<bookmark xml-lang="en-US" branch="index" id="bm_id1744743"><bookmark_value>calculating;regression curves</bookmark_value>
<bookmark_value>regression curves in charts</bookmark_value>
<bookmark_value>trend lines in charts</bookmark_value>
<bookmark_value>mean value lines in charts</bookmark_value>
</bookmark>
<bookmark xml-lang="en-US" branch="hid/.uno:InsertMenuTrendlines" id="bm_id2001709" localize="false"/>
<paragraph xml-lang="en-US" id="hd_id5409405" role="heading" level="1" l10n="CHG"><variable id="regression"><link href="text/schart/01/04050100.xhp">Trend Lines</link>
</variable></paragraph>
<paragraph xml-lang="en-US" id="par_id7272255" role="paragraph" l10n="CHG"><variable id="trendlinestext"><ahelp hid=".">Regression curves, also known as trend lines, can be added to all 2D chart types except for Pie and Stock charts.</ahelp>
</variable></paragraph>
<embed href="text/schart/01/04030000.xhp#context"/>
<section id="howtoget">
<embed href="text/schart/00/00000004.xhp#trendlines"/>
</section><comment>None</comment>
<bookmark xml-lang="en-US" branch="hid/HID_SCH_TRENDLINE_RB_NONE" id="bm_id2659077" localize="false"/>
<paragraph xml-lang="en-US" id="par_id143436" role="paragraph" l10n="NEW"><ahelp hid="." visibility="hidden">No trend line is shown.</ahelp></paragraph><comment>Linear</comment>
<bookmark xml-lang="en-US" branch="hid/HID_SCH_TRENDLINE_RB_LINEAR" id="bm_id6944181" localize="false"/>
<paragraph xml-lang="en-US" id="par_id5716727" role="paragraph" l10n="NEW"><ahelp hid="." visibility="hidden">A linear trend line is shown.</ahelp></paragraph><comment>Logarithmic</comment>
<bookmark xml-lang="en-US" branch="hid/HID_SCH_TRENDLINE_RB_LOGARITHMIC" id="bm_id886257" localize="false"/>
<paragraph xml-lang="en-US" id="par_id5840021" role="paragraph" l10n="NEW"><ahelp hid="." visibility="hidden">A logarithmic trend line is shown.</ahelp></paragraph><comment>Exponential</comment>
<bookmark xml-lang="en-US" branch="hid/HID_SCH_TRENDLINE_RB_EXPONENTIAL" id="bm_id8892483" localize="false"/>
<paragraph xml-lang="en-US" id="par_id9417096" role="paragraph" l10n="NEW"><ahelp hid="." visibility="hidden">An exponential trend line is shown.</ahelp></paragraph><comment>Power</comment>
<bookmark xml-lang="en-US" branch="hid/HID_SCH_TRENDLINE_RB_POWER" id="bm_id7198400" localize="false"/>
<paragraph xml-lang="en-US" id="par_id8482924" role="paragraph" l10n="NEW"><ahelp hid="." visibility="hidden">A power trend line is shown.</ahelp></paragraph><comment>Show equation</comment>
<bookmark xml-lang="en-US" branch="hid/HID_SCH_TRENDLINE_SHOW_EQUATION" id="bm_id4724875" localize="false"/>
<paragraph xml-lang="en-US" id="par_id8962370" role="paragraph" l10n="NEW"><ahelp hid="." visibility="hidden">Shows the trend line equation next to the trend line.</ahelp></paragraph><comment>Show correlation coefficient (R2)</comment>
<bookmark xml-lang="en-US" branch="hid/HID_SCH_TRENDLINE_SHOW_R_SQUARED" id="bm_id9068880" localize="false"/>
<paragraph xml-lang="en-US" id="par_id6889858" role="paragraph" l10n="CHG"><ahelp hid="." visibility="hidden">Shows the coefficient of determination next to the trend line.</ahelp></paragraph>
<paragraph xml-lang="en-US" id="par_id8398998" role="note" l10n="CHG">If you insert a trend line to a chart type that uses categories, like <emph>Line </emph>or <emph>Column, </emph>then the numbers 1, 2, 3, <emph>…</emph> are used as x-values to calculate the trend line.</paragraph>
<list type="ordered">
<listitem>
<paragraph xml-lang="en-US" id="par_id5676747" role="paragraph" l10n="CHG">To insert trend lines for all data series, double-click the chart to enter edit mode. Choose <item type="menuitem">Insert - Trend Lines</item>, then select the type of trend line from None, Linear, Logarithmic, Exponential, or Power trend line.</paragraph>
</listitem>
<listitem>
<paragraph xml-lang="en-US" id="par_id4349192" role="paragraph" l10n="CHG">To insert a trend line for a single data series, select the data series in the chart, right-click to open the context menu, and choose <item type="menuitem">Insert - Trend Line</item>.</paragraph>
</listitem>
<listitem>
<paragraph xml-lang="en-US" id="par_id9337443" role="paragraph" l10n="NEW">To delete a single trend line or mean value line, click the line, then press the Del key.</paragraph>
</listitem>
<listitem>
<paragraph xml-lang="en-US" id="par_id4529251" role="paragraph" l10n="NEW">To delete all trend lines, choose <item type="menuitem">Insert - Trend Lines</item>, then select <emph>None</emph>.</paragraph>
</listitem>
</list>
<paragraph xml-lang="en-US" id="par_id296334" role="note" l10n="CHG">A trend line is shown in the legend automatically.</paragraph>
<bookmark xml-lang="en-US" branch="hid/.uno:InsertMenuMeanValues" id="bm_id7854243" localize="false"/>
<paragraph xml-lang="en-US" id="par_id4072084" role="paragraph" l10n="NEW"><ahelp hid=".">Mean Value Lines are special trend lines that show the mean value. Use <item type="menuitem">Insert - Mean Value Lines</item> to insert mean value lines for data series.</ahelp></paragraph>
<embed href="text/schart/01/04030000.xhp#context"/>
<paragraph xml-lang="en-US" id="par_id9569689" role="paragraph" l10n="NEW">The trend line has the same color as the corresponding data series. To change the line properties, select the trend line and choose <item type="menuitem">Format - Format Selection - Line</item>.</paragraph>
<bookmark xml-lang="en-US" branch="hid/.uno:InsertTrendlineEquation" id="bm_id2754600" localize="false"/>
<paragraph xml-lang="en-US" id="par_id846888" role="paragraph" l10n="NEW"><ahelp hid=".">To show the trend line equation, select the trend line in the chart, right-click to open the context menu, and choose <emph>Insert Trend Line Equation</emph>.</ahelp></paragraph>
<paragraph xml-lang="en-US" id="par_id8962065" role="paragraph" l10n="CHG">When the chart is in edit mode, %PRODUCTNAME gives you the equation of the trend line and the coefficient of determination R². Click on the trend line to see the information in the status bar.</paragraph>
<paragraph xml-lang="en-US" id="par_id1328470" role="note" l10n="NEW">For a category chart (for example a line chart), the regression information is calculated using numbers 1, 2, 3, … as x-values. This is also true if your data series uses other numbers as names for the x-values. For such charts the XY chart type might be more suitable.</paragraph>
<paragraph xml-lang="en-US" id="par_id8092593" role="paragraph" l10n="CHG">To show the equation and the coefficient of determination, select the regression curve and choose <item type="menuitem">Format - Format Selection - Equation</item>. <comment>see http://specs.openoffice.org/chart/DisplayTrendLineEquations.odt</comment></paragraph><comment>hid</comment>
<paragraph xml-lang="en-US" id="par_id7971434" role="paragraph" l10n="NEW"><ahelp hid="." visibility="hidden">Enable Show equation to see the equation of the regression curve.</ahelp></paragraph><comment>hid</comment>
<paragraph xml-lang="en-US" id="par_id558793" role="paragraph" l10n="CHG"><ahelp hid="." visibility="hidden">Enable Show Coefficient of Determination to see the determination coefficient of the regression curve.</ahelp></paragraph>
<paragraph xml-lang="en-US" id="par_id7735221" role="paragraph" l10n="NEW">You can also calculate the parameters using Calc functions as follows.</paragraph>
<paragraph xml-lang="en-US" id="hd_id5744193" role="heading" level="1" l10n="NEW">The linear regression equation</paragraph>
<paragraph xml-lang="en-US" id="par_id9251991" role="paragraph" l10n="NEW">The <emph>linear regression</emph> follows the equation <item type="literal">y=m*x+b</item>.</paragraph>
<paragraph xml-lang="en-US" id="par_id7951902" role="code" l10n="NEW">m = SLOPE(Data_Y;Data_X) </paragraph>
<paragraph xml-lang="en-US" id="par_id6637165" role="code" l10n="NEW">b = INTERCEPT(Data_Y ;Data_X) </paragraph>
<paragraph xml-lang="en-US" id="par_id7879268" role="paragraph" l10n="NEW">Calculate the coefficient of determination by</paragraph>
<paragraph xml-lang="en-US" id="par_id9244361" role="code" l10n="NEW">r² = RSQ(Data_Y;Data_X) </paragraph>
<paragraph xml-lang="en-US" id="par_id2083498" role="paragraph" l10n="NEW">Besides m, b and r² the array function <emph>LINEST</emph> provides additional statistics for a regression analysis.</paragraph>
<paragraph xml-lang="en-US" id="hd_id2538834" role="heading" level="1" l10n="NEW">The logarithm regression equation</paragraph>
<paragraph xml-lang="en-US" id="par_id394299" role="paragraph" l10n="NEW">The <emph>logarithm regression</emph> follows the equation <item type="literal">y=a*ln(x)+b</item>.</paragraph>
<paragraph xml-lang="en-US" id="par_id2134159" role="code" l10n="NEW">a = SLOPE(Data_Y;LN(Data_X)) </paragraph>
<paragraph xml-lang="en-US" id="par_id5946531" role="code" l10n="NEW">b = INTERCEPT(Data_Y ;LN(Data_X)) </paragraph>
<paragraph xml-lang="en-US" id="par_id5649281" role="code" l10n="NEW">r² = RSQ(Data_Y;LN(Data_X)) </paragraph>
<paragraph xml-lang="en-US" id="hd_id7874080" role="heading" level="1" l10n="NEW">The exponential regression equation</paragraph>
<paragraph xml-lang="en-US" id="par_id4679097" role="paragraph" l10n="NEW"> For exponential regression curves a transformation to a linear model takes place. The optimal curve fitting is related to the linear model and the results are interpreted accordingly. </paragraph>
<paragraph xml-lang="en-US" id="par_id9112216" role="paragraph" l10n="NEW">The exponential regression follows the equation <item type="literal">y=b*exp(a*x)</item> or <item type="literal">y=b*m^x</item>, which is transformed to <item type="literal">ln(y)=ln(b)+a*x</item> or <item type="literal">ln(y)=ln(b)+ln(m)*x</item> respectively.</paragraph>
<paragraph xml-lang="en-US" id="par_id4416638" role="code" l10n="NEW">a = SLOPE(LN(Data_Y);Data_X) </paragraph>
<paragraph xml-lang="en-US" id="par_id1039155" role="paragraph" l10n="CHG">The variables for the second variation are calculated as follows:</paragraph>
<paragraph xml-lang="en-US" id="par_id7184057" role="code" l10n="NEW">m = EXP(SLOPE(LN(Data_Y);Data_X)) </paragraph>
<paragraph xml-lang="en-US" id="par_id786767" role="code" l10n="NEW">b = EXP(INTERCEPT(LN(Data_Y);Data_X)) </paragraph>
<paragraph xml-lang="en-US" id="par_id7127292" role="paragraph" l10n="NEW">Calculate the coefficient of determination by</paragraph>
<paragraph xml-lang="en-US" id="par_id5437177" role="code" l10n="NEW">r² = RSQ(LN(Data_Y);Data_X) </paragraph>
<paragraph xml-lang="en-US" id="par_id6946317" role="paragraph" l10n="NEW">Besides m, b and r² the array function LOGEST provides additional statistics for a regression analysis.</paragraph>
<paragraph xml-lang="en-US" id="hd_id6349375" role="heading" level="1" l10n="NEW">The power regression equation</paragraph>
<paragraph xml-lang="en-US" id="par_id1857661" role="paragraph" l10n="NEW"> For <emph>power regression</emph> curves a transformation to a linear model takes place. The power regression follows the equation <item type="literal">y=b*x^a</item> , which is transformed to <item type="literal">ln(y)=ln(b)+a*ln(x)</item>.</paragraph>
<paragraph xml-lang="en-US" id="par_id8517105" role="code" l10n="NEW">a = SLOPE(LN(Data_Y);LN(Data_X)) </paragraph>
<paragraph xml-lang="en-US" id="par_id9827265" role="code" l10n="NEW">b = EXP(INTERCEPT(LN(Data_Y);LN(Data_X)) </paragraph>
<paragraph xml-lang="en-US" id="par_id2357249" role="code" l10n="NEW">r² = RSQ(LN(Data_Y);LN(Data_X)) </paragraph>
<paragraph xml-lang="en-US" id="hd_id9204077" role="heading" level="1" l10n="NEW">Constraints<comment>UFI: is this still so?</comment></paragraph>
<paragraph xml-lang="en-US" id="par_id7393719" role="paragraph" l10n="CHG"> The calculation of the trend line considers only data pairs with the following values:</paragraph>
<list type="ordered">
<listitem>
<paragraph xml-lang="en-US" id="par_id7212744" role="paragraph" l10n="NEW">logarithm regression: only positive x-values are considered,</paragraph>
</listitem>
<listitem>
<paragraph xml-lang="en-US" id="par_id1664479" role="paragraph" l10n="NEW">exponential regression: only positive y-values are considered,</paragraph>
</listitem>
<listitem>
<paragraph xml-lang="en-US" id="par_id8734702" role="paragraph" l10n="NEW">power regression: only positive x-values and positive y-values are considered.</paragraph>
</listitem>
</list>
<paragraph xml-lang="en-US" id="par_id181279" role="paragraph" l10n="NEW">You should transform your data accordingly; it is best to work on a copy of the original data and transform the copied data.</paragraph>
<paragraph xml-lang="en-US" id="hd_id7907040" role="heading" level="1" l10n="NEW">The polynomial regression equation</paragraph>
<paragraph xml-lang="en-US" id="par_id8918729" role="paragraph" l10n="NEW">A <emph>polynomial regression</emph> curve cannot be added automatically. You must calculate this curve manually. </paragraph>
<paragraph xml-lang="en-US" id="par_id33875" role="paragraph" l10n="NEW">Create a table with the columns x, x², x³, … , xⁿ, y up to the desired degree n. </paragraph>
<paragraph xml-lang="en-US" id="par_id8720053" role="paragraph" l10n="NEW">Use the formula <item type="literal">=LINEST(Data_Y,Data_X)</item> with the complete range x to xⁿ (without headings) as Data_X. </paragraph>
<paragraph xml-lang="en-US" id="par_id5068514" role="paragraph" l10n="NEW">The first row of the LINEST output contains the coefficients of the regression polynomial, with the coefficient of xⁿ at the leftmost position.</paragraph>
<paragraph xml-lang="en-US" id="par_id8202154" role="paragraph" l10n="NEW">The first element of the third row of the LINEST output is the value of r². See the <link href="text/scalc/01/04060107.xhp#Section8">LINEST</link> function for details on proper use and an explanation of the other output parameters.</paragraph>
<section id="relatedtopics">
<paragraph xml-lang="en-US" id="par_id4562211" role="paragraph" l10n="CHG"><link href="text/schart/01/04050000.xhp">Y Error Bars tab page</link></paragraph>
</section>
</body>
</helpdocument>
|
