<?xml version="1.0" encoding="UTF-8"?>
<rss xmlns:content="http://purl.org/rss/1.0/modules/content/" xmlns:dc="http://purl.org/dc/elements/1.1/" xmlns:rdf="http://www.w3.org/1999/02/22-rdf-syntax-ns#" xmlns:taxo="http://purl.org/rss/1.0/modules/taxonomy/" version="2.0">
  <channel>
    <title>topic Re: SAS loses decimal accuracy in SAS Enterprise Guide</title>
    <link>https://communities.sas.com/t5/SAS-Enterprise-Guide/SAS-loses-decimal-accuracy/m-p/407747#M26166</link>
    <description>&lt;P&gt;To the extent you can work with integers instead of decimal fractions, the results will improve.&amp;nbsp; If it would be practical, try:&lt;/P&gt;
&lt;P&gt;&amp;nbsp;&lt;/P&gt;
&lt;UL&gt;
&lt;LI&gt;Multiply costs * 1000&lt;/LI&gt;
&lt;LI&gt;Sum up the new costs&lt;/LI&gt;
&lt;LI&gt;Divide the totals by 1000&lt;/LI&gt;
&lt;/UL&gt;
&lt;P&gt;Those totals should be closer.&lt;/P&gt;</description>
    <pubDate>Thu, 26 Oct 2017 17:33:57 GMT</pubDate>
    <dc:creator>Astounding</dc:creator>
    <dc:date>2017-10-26T17:33:57Z</dc:date>
    <item>
      <title>SAS loses decimal accuracy</title>
      <link>https://communities.sas.com/t5/SAS-Enterprise-Guide/SAS-loses-decimal-accuracy/m-p/407713#M26160</link>
      <description>&lt;P&gt;Hi&lt;BR /&gt;&lt;BR /&gt;I have a weird problem. I am operating on numbers which add up to around 800 000 000. I have a dataset which I need to split into minidatasets with 1000 000 records. The problem is that when I split my dataset, I loose decimal precision, for example:&lt;BR /&gt;&lt;BR /&gt;Befote splitting the sum of my costs was 810268311.8536920 &lt;BR /&gt;&lt;BR /&gt;After splitting - 810268311.8401380&lt;BR /&gt;&lt;BR /&gt;Splitted datasets have the total number of records equal to number of records which was in the initial table. &lt;BR /&gt;&lt;BR /&gt;What can I do about that?&lt;BR /&gt;&lt;BR /&gt;Thank you in advance!&lt;/P&gt;</description>
      <pubDate>Thu, 26 Oct 2017 16:33:56 GMT</pubDate>
      <guid>https://communities.sas.com/t5/SAS-Enterprise-Guide/SAS-loses-decimal-accuracy/m-p/407713#M26160</guid>
      <dc:creator>Dontik</dc:creator>
      <dc:date>2017-10-26T16:33:56Z</dc:date>
    </item>
    <item>
      <title>Re: SAS loses decimal accuracy</title>
      <link>https://communities.sas.com/t5/SAS-Enterprise-Guide/SAS-loses-decimal-accuracy/m-p/407717#M26161</link>
      <description>&lt;P&gt;Your numbers never had the accuracy. The 8 byte floating point numeric format used by SAS can only hold up to fifteen significant digits accurately. Your numbers have sixteen digits; you will lose precision on the sixteenth and above digits.&lt;/P&gt;
&lt;P&gt;&amp;nbsp;&lt;/P&gt;
&lt;P&gt;Tom&lt;/P&gt;</description>
      <pubDate>Thu, 26 Oct 2017 16:36:29 GMT</pubDate>
      <guid>https://communities.sas.com/t5/SAS-Enterprise-Guide/SAS-loses-decimal-accuracy/m-p/407717#M26161</guid>
      <dc:creator>TomKari</dc:creator>
      <dc:date>2017-10-26T16:36:29Z</dc:date>
    </item>
    <item>
      <title>Re: SAS loses decimal accuracy</title>
      <link>https://communities.sas.com/t5/SAS-Enterprise-Guide/SAS-loses-decimal-accuracy/m-p/407720#M26162</link>
      <description>&lt;P&gt;Hmm so what is there to do about that? Can I at least keep the number at 0.01 decimal precision? 810268311.85 - so that I could get this exact number after splitting, I do not need anything more precise. I"ve tried adding round(cost,0.01) to my data step which splits the data but that didn't help.&lt;/P&gt;</description>
      <pubDate>Thu, 26 Oct 2017 16:51:13 GMT</pubDate>
      <guid>https://communities.sas.com/t5/SAS-Enterprise-Guide/SAS-loses-decimal-accuracy/m-p/407720#M26162</guid>
      <dc:creator>Dontik</dc:creator>
      <dc:date>2017-10-26T16:51:13Z</dc:date>
    </item>
    <item>
      <title>Re: SAS loses decimal accuracy</title>
      <link>https://communities.sas.com/t5/SAS-Enterprise-Guide/SAS-loses-decimal-accuracy/m-p/407730#M26163</link>
      <description>&lt;P&gt;There is an exhausting treatise on this topic here:&lt;/P&gt;
&lt;UL&gt;
&lt;LI&gt;&lt;A href="https://blogs.sas.com/content/sasdummy/2012/03/01/precision-in-sas-numbers/" target="_self"&gt;Numerical precision in SAS&lt;/A&gt;&lt;/LI&gt;
&lt;/UL&gt;
&lt;P&gt;The behavior of SAS is the same as any software that relies on floating point math, defined by an IEEE standard.&lt;/P&gt;
&lt;P&gt;&amp;nbsp;&lt;/P&gt;
&lt;P&gt;There is a newer &lt;A href="http://go.documentation.sas.com/?docsetId=lesysoptsref&amp;amp;docsetTarget=n130tap9xs8q36n14gq23sci5y52.htm&amp;amp;docsetVersion=9.4&amp;amp;locale=en" target="_self"&gt;DECIMALCONV= option that might help&lt;/A&gt;.&amp;nbsp; From the doc:&lt;/P&gt;
&lt;P&gt;&amp;nbsp;&lt;/P&gt;
&lt;H4 class="xis-argument"&gt;DECIMALCONV=STDIEEE&lt;/H4&gt;
&lt;DIV class="xis-argumentDescription"&gt;
&lt;P class="xis-paraSimpleFirst"&gt;specifies to convert and format decimal values by using the IEEE Standard for Floating-Point Arithmetic 754-2008. Using the STDIEEE argument improves the accuracy and readability of floating-point numbers. In some cases, more significant digits can be displayed in the same field width.&lt;/P&gt;
&lt;/DIV&gt;</description>
      <pubDate>Thu, 26 Oct 2017 17:01:51 GMT</pubDate>
      <guid>https://communities.sas.com/t5/SAS-Enterprise-Guide/SAS-loses-decimal-accuracy/m-p/407730#M26163</guid>
      <dc:creator>ChrisHemedinger</dc:creator>
      <dc:date>2017-10-26T17:01:51Z</dc:date>
    </item>
    <item>
      <title>Re: SAS loses decimal accuracy</title>
      <link>https://communities.sas.com/t5/SAS-Enterprise-Guide/SAS-loses-decimal-accuracy/m-p/407737#M26164</link>
      <description>Unfortunately, my SAS says "unrecognized SAS option name, DECIMALCONV" &lt;span class="lia-unicode-emoji" title=":confused_face:"&gt;😕&lt;/span&gt;</description>
      <pubDate>Thu, 26 Oct 2017 17:15:40 GMT</pubDate>
      <guid>https://communities.sas.com/t5/SAS-Enterprise-Guide/SAS-loses-decimal-accuracy/m-p/407737#M26164</guid>
      <dc:creator>Dontik</dc:creator>
      <dc:date>2017-10-26T17:15:40Z</dc:date>
    </item>
    <item>
      <title>Re: SAS loses decimal accuracy</title>
      <link>https://communities.sas.com/t5/SAS-Enterprise-Guide/SAS-loses-decimal-accuracy/m-p/407741#M26165</link>
      <description>&lt;P&gt;DECIMALCONV= was added in SAS 9.4.&amp;nbsp;&lt;/P&gt;</description>
      <pubDate>Thu, 26 Oct 2017 17:19:53 GMT</pubDate>
      <guid>https://communities.sas.com/t5/SAS-Enterprise-Guide/SAS-loses-decimal-accuracy/m-p/407741#M26165</guid>
      <dc:creator>ChrisHemedinger</dc:creator>
      <dc:date>2017-10-26T17:19:53Z</dc:date>
    </item>
    <item>
      <title>Re: SAS loses decimal accuracy</title>
      <link>https://communities.sas.com/t5/SAS-Enterprise-Guide/SAS-loses-decimal-accuracy/m-p/407747#M26166</link>
      <description>&lt;P&gt;To the extent you can work with integers instead of decimal fractions, the results will improve.&amp;nbsp; If it would be practical, try:&lt;/P&gt;
&lt;P&gt;&amp;nbsp;&lt;/P&gt;
&lt;UL&gt;
&lt;LI&gt;Multiply costs * 1000&lt;/LI&gt;
&lt;LI&gt;Sum up the new costs&lt;/LI&gt;
&lt;LI&gt;Divide the totals by 1000&lt;/LI&gt;
&lt;/UL&gt;
&lt;P&gt;Those totals should be closer.&lt;/P&gt;</description>
      <pubDate>Thu, 26 Oct 2017 17:33:57 GMT</pubDate>
      <guid>https://communities.sas.com/t5/SAS-Enterprise-Guide/SAS-loses-decimal-accuracy/m-p/407747#M26166</guid>
      <dc:creator>Astounding</dc:creator>
      <dc:date>2017-10-26T17:33:57Z</dc:date>
    </item>
  </channel>
</rss>

