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
|
<?xml version="1.0" encoding="UTF-8"?>
<helpdocument version="1.0">
<!--
* This file is part of the LibreOffice project.
*
* This Source Code Form is subject to the terms of the Mozilla Public
* License, v. 2.0. If a copy of the MPL was not distributed with this
* file, You can obtain one at http://mozilla.org/MPL/2.0/.
*
-->
<meta>
<topic id="textscalc01func_forecastetspimultxml">
<title id="tit" xml-lang="en-US">FORECAST.ETS.PI.MULT</title>
<filename>/text/scalc/01/func_forecastetspimult.xhp</filename>
</topic>
</meta>
<body>
<section id="forecastetspimult">
<bookmark xml-lang="en-US" branch="hid/SC_HID_FUNC_FORECAST_ETS_PIM" id="bm_id0603201617144585" localize="false"/>
<bookmark xml-lang="en-US" branch="index" id="bm_id976559765597655">
<bookmark_value>FORECAST.ETS.PI.MULT function</bookmark_value>
</bookmark>
<paragraph id="hd_id0603201617134175" role="heading" level="1" xml-lang="en-US"><link href="text/scalc/01/func_forecastetspimult.xhp">FORECAST.ETS.PI.MULT function</link></paragraph>
<paragraph id="par_id0603201617141750" role="paragraph" xml-lang="en-US"><ahelp hid="HID_FUNC_FORECAST_ETS_PIM">Calculates Prediction Interval(s) based on the historical data using ETS or EDS algorithms.</ahelp>. EDS is used when argument <emph>period_length</emph> is 0, otherwise ETS is used.</paragraph>
<embed href="text/scalc/01/exponsmooth_embd.xhp#intro"/>
<paragraph id="par_id0603201610005998" role="paragraph" xml-lang="en-US">FORECAST.ETS.PI.MULT calculates with the model</paragraph>
<embed href="text/scalc/01/exponsmooth_embd.xhp#etsmult"/>
<paragraph id="hd_id0603201610005973" role="heading" level="2" xml-lang="en-US">Syntax</paragraph>
<paragraph id="par_id0603201610010044" role="code" xml-lang="en-US">FORECAST.ETS.PI.MULT(target, values, timeline, [confidence], [period_length], [completion], [aggregation])</paragraph>
<embed href="text/scalc/01/exponsmooth_embd.xhp#target"/>
<embed href="text/scalc/01/exponsmooth_embd.xhp#values"/>
<embed href="text/scalc/01/exponsmooth_embd.xhp#timeline"/>
<embed href="text/scalc/01/exponsmooth_embd.xhp#confidence"/>
<embed href="text/scalc/01/exponsmooth_embd.xhp#numsampperiod"/>
<embed href="text/scalc/01/exponsmooth_embd.xhp#datacompletion"/>
<embed href="text/scalc/01/exponsmooth_embd.xhp#aggregation"/>
<paragraph id="par_id0403201618595126" role="paragraph" xml-lang="en-US">For example, with a 90% Confidence level, a 90% prediction interval will be computed (90% of future points are to fall within this radius from forecast). </paragraph>
<paragraph id="par_id0403201618595143" role="note" xml-lang="en-US">Note on Prediction Intervals: there is no exact mathematical way to calculate this for forecasts, there are various approximations. Prediction Intervals tend to be increasingly 'over-optimistic' with increasing distance of the forecast-X to the observation data set.</paragraph>
<paragraph id="par_id0403201618595150" role="paragraph" xml-lang="en-US">For ETS, Calc uses an approximation based on 1000 calculations with random variations within the standard deviation of the observation data set (the historical values).</paragraph>
<embed href="text/scalc/01/exponsmooth_embd.xhp#exampledata"/>
<paragraph id="hd_id04032016185123" role="code" xml-lang="en-US">=FORECAST.ETS.PI.MULT(DATE(2014;1;1);Values;Timeline;0,9;1;TRUE();1)</paragraph>
<paragraph id="hd_id04032016112394554" role="paragraph" xml-lang="en-US">Returns 20.1040952101013, the multiplicative prediction interval forecast for January 2014 based on <emph>Values</emph> and <emph>Timeline</emph> named ranges above, with one sample per period, no missing data, and AVERAGE as aggregation.</paragraph>
<paragraph id="hd_id04032123185123" role="code" xml-lang="en-US">=FORECAST.ETS.PI.MULT(DATE(2014;1;1);Values;Timeline;0.8;4;TRUE();7)</paragraph>
<paragraph id="hd_id040312316112394554" role="paragraph" xml-lang="en-US">Returns 27.5285874381574, the multiplicative prediction interval forecast for January 2014 based on <emph>Values</emph> and <emph>Timeline</emph> named ranges above, with confidence level of 0.8, period length of 4, no missing data, and SUM as aggregation.</paragraph>
</section>
<section id="relatedtopics">
<paragraph id="par_id0603201619261276" role="paragraph" xml-lang="en-US">See also:
<link href="text/scalc/01/func_forecastetsadd.xhp">FORECAST.ETS.ADD</link>,
<link href="text/scalc/01/func_forecastetsmult.xhp">FORECAST.ETS.MULT</link>,
<link href="text/scalc/01/func_forecastetsstatadd.xhp">FORECAST.ETS.STAT.ADD</link>,
<link href="text/scalc/01/func_forecastetsstatmult.xhp">FORECAST.ETS.STAT.MULT</link>,
<link href="text/scalc/01/func_forecastetspiadd.xhp">FORECAST.ETS.PI.ADD</link>,
<link href="text/scalc/01/func_forecastetsseason.xhp">FORECAST.ETS.SEASONALITY</link>,
<link href="text/scalc/01/04060185.xhp#forecast">FORECAST</link>,
<link href="text/scalc/01/04060185.xhp#forecastlinear">FORECAST.LINEAR</link>
</paragraph>
</section>
</body>
</helpdocument>
|
