{"id":922,"date":"2025-06-20T17:22:54","date_gmt":"2025-06-20T17:22:54","guid":{"rendered":"https:\/\/content.one.lumenlearning.com\/calculus2\/?post_type=chapter&#038;p=922"},"modified":"2025-09-11T16:41:45","modified_gmt":"2025-09-11T16:41:45","slug":"introduction-to-power-series-learn-it-2","status":"publish","type":"chapter","link":"https:\/\/content.one.lumenlearning.com\/calculus2\/chapter\/introduction-to-power-series-learn-it-2\/","title":{"raw":"Introduction to Power Series: Learn It 2","rendered":"Introduction to Power Series: Learn It 2"},"content":{"raw":"<h2 data-type=\"title\">Convergence of a Power Series<\/h2>\r\n<p class=\"whitespace-normal break-words\">Now we need to address a crucial question: for which values of [latex]x[\/latex] does a power series actually converge? Unlike regular series with constant terms, power series behavior depends entirely on the value of [latex]x[\/latex]. The series might converge for some [latex]x[\/latex]-values and diverge for others.<\/p>\r\n\r\n<section class=\"textbox tryIt\" aria-label=\"Try It\">[ohm_question hide_question_numbers=1]311402[\/ohm_question]<\/section>\r\n<p class=\"whitespace-normal break-words\">Every power series has at least one guaranteed convergence point: <strong>its center<\/strong>.<\/p>\r\n<p class=\"whitespace-pre-wrap break-words\">For a power series [latex]\\displaystyle\\sum_{n=0}^{\\infty} c_n (x-a)^n[\/latex] centered at [latex]x = a[\/latex], when we substitute [latex]x = a[\/latex]:<\/p>\r\n<p class=\"whitespace-pre-wrap break-words\" style=\"text-align: center;\">[latex]\\displaystyle\\sum_{n=0}^{\\infty} c_n (a-a)^n = c_0 + 0 + 0 + \\cdots = c_0[\/latex]<\/p>\r\n<p class=\"whitespace-normal break-words\">The series reduces to just the constant term [latex]c_0[\/latex], so it always converges at its center.<\/p>\r\n<p class=\"whitespace-normal break-words\">Beyond the center, a power series can behave in exactly one of three ways.<\/p>\r\n\r\n<section class=\"textbox keyTakeaway\" aria-label=\"Key Takeaway\">\r\n<h3>convergence theorem for power series<\/h3>\r\n<p class=\"whitespace-normal break-words\">For any power series [latex]\\displaystyle\\sum_{n=0}^{\\infty} c_n (x-a)^n[\/latex], exactly one of these occurs:<\/p>\r\n\r\n<ol class=\"[&amp;:not(:last-child)_ul]:pb-1 [&amp;:not(:last-child)_ol]:pb-1 list-decimal space-y-1.5 pl-7\">\r\n \t<li class=\"whitespace-normal break-words\"><strong>Converges only at the center:<\/strong> The series converges at [latex]x = a[\/latex] and diverges everywhere else<\/li>\r\n \t<li class=\"whitespace-normal break-words\"><strong>Converges everywhere:<\/strong> The series converges for all real numbers [latex]x[\/latex]<\/li>\r\n \t<li class=\"whitespace-normal break-words\"><strong>Converges in an interval:<\/strong> There exists [latex]R &gt; 0[\/latex] such that:\r\n<ul class=\"[&amp;:not(:last-child)_ul]:pb-1 [&amp;:not(:last-child)_ol]:pb-1 list-disc space-y-1.5 pl-7\">\r\n \t<li class=\"whitespace-normal break-words\">The series converges when [latex]|x-a| &lt; R[\/latex]<\/li>\r\n \t<li class=\"whitespace-normal break-words\">The series diverges when [latex]|x-a| &gt; R[\/latex]<\/li>\r\n \t<li class=\"whitespace-normal break-words\">At the boundary points where [latex]|x-a| = R[\/latex], the series may converge or diverge<\/li>\r\n<\/ul>\r\n<\/li>\r\n<\/ol>\r\n<\/section><section class=\"textbox connectIt\">\r\n<p style=\"text-align: center;\"><strong>Proof<\/strong><\/p>\r\n\r\n\r\n<hr \/>\r\n<p id=\"fs-id1170572107834\">Suppose that the power series is centered at [latex]a=0[\/latex]. (For a series centered at a value of <em data-effect=\"italics\">a<\/em> other than zero, the result follows by letting [latex]y=x-a[\/latex] and considering the series [latex]{\\displaystyle\\sum _{n=1}^{\\infty}} {c}_{n} {y}^{n}[\/latex].) We must first prove the following fact:<\/p>\r\n<p id=\"fs-id1170572109983\">If there exists a real number [latex]d\\ne 0[\/latex] such that [latex]\\displaystyle\\sum _{n=0}^{\\infty }{c}_{n}{d}^{n}[\/latex] converges, then the series [latex]\\displaystyle\\sum _{n=0}^{\\infty }{c}_{n}{x}^{n}[\/latex] converges absolutely for all <em data-effect=\"italics\">x<\/em> such that [latex]|x|&lt;|d|[\/latex].<\/p>\r\n<p id=\"fs-id1170572366747\">Since [latex]\\displaystyle\\sum _{n=0}^{\\infty }{c}_{n}{d}^{n}[\/latex] converges, the <em data-effect=\"italics\">n<\/em>th term [latex]{c}_{n}{d}^{n}\\to 0[\/latex] as [latex]n\\to \\infty [\/latex]. Therefore, there exists an integer <em data-effect=\"italics\">N<\/em> such that [latex]|{c}_{n}{d}^{n}|\\le 1[\/latex] for all [latex]n\\ge N[\/latex]. Writing<\/p>\r\n\r\n<div id=\"fs-id1170571538405\" class=\"unnumbered\" style=\"text-align: center;\" data-type=\"equation\" data-label=\"\">[latex]|{c}_{n}{x}^{n}|=|{c}_{n}{d}^{n}|{|\\frac{x}{d}|}^{n}[\/latex],<\/div>\r\n&nbsp;\r\n<p id=\"fs-id1170571637176\">we conclude that, for all [latex]n\\ge N[\/latex],<\/p>\r\n\r\n<div id=\"fs-id1170572436828\" class=\"unnumbered\" style=\"text-align: center;\" data-type=\"equation\" data-label=\"\">[latex]|{c}_{n}{x}^{n}|\\le {|\\frac{x}{d}|}^{n}[\/latex].<\/div>\r\n&nbsp;\r\n<p id=\"fs-id1170571611376\">The series<\/p>\r\n\r\n<div id=\"fs-id1170572375703\" class=\"unnumbered\" style=\"text-align: center;\" data-type=\"equation\" data-label=\"\">[latex]\\displaystyle\\sum _{n=N}^{\\infty }{|\\frac{x}{d}|}^{n}[\/latex]<\/div>\r\n&nbsp;\r\n<p id=\"fs-id1170571653945\">is a geometric series that converges if [latex]|\\frac{x}{d}|&lt;1[\/latex]. Therefore, by the comparison test, we conclude that [latex]\\displaystyle\\sum _{n=N}^{\\infty }{c}_{n}{x}^{n}[\/latex] also converges for [latex]|x|&lt;|d|[\/latex]. Since we can add a finite number of terms to a convergent series, we conclude that [latex]\\displaystyle\\sum _{n=0}^{\\infty }{c}_{n}{x}^{n}[\/latex] converges for [latex]|x|&lt;|d|[\/latex].<\/p>\r\n<p id=\"fs-id1170571595417\">With this result, we can now prove the theorem. Consider the series<\/p>\r\n\r\n<div id=\"fs-id1170572250450\" class=\"unnumbered\" style=\"text-align: center;\" data-type=\"equation\" data-label=\"\">[latex]\\displaystyle\\sum _{n=0}^{\\infty }{a}_{n}{x}^{n}[\/latex]<\/div>\r\n&nbsp;\r\n<p id=\"fs-id1170572224801\">and let <em data-effect=\"italics\">S<\/em> be the set of real numbers for which the series converges. Suppose that the set [latex]S=\\left\\{0\\right\\}[\/latex]. Then the series falls under case i. Suppose that the set <em data-effect=\"italics\">S<\/em> is the set of all real numbers. Then the series falls under case ii. Suppose that [latex]S\\ne \\left\\{0\\right\\}[\/latex] and <em data-effect=\"italics\">S<\/em> is not the set of real numbers. Then there exists a real number [latex]x*\\ne 0[\/latex] such that the series does not converge. Thus, the series cannot converge for any <em data-effect=\"italics\">x<\/em> such that [latex]|x|&gt;|x*|[\/latex]. Therefore, the set <em data-effect=\"italics\">S<\/em> must be a bounded set, which means that it must have a smallest upper bound. (This fact follows from the Least Upper Bound Property for the real numbers, which is beyond the scope of this text and is covered in real analysis courses.) Call that smallest upper bound <em data-effect=\"italics\">R<\/em>. Since [latex]S\\ne \\left\\{0\\right\\}[\/latex], the number [latex]R &lt; 0[\/latex]. Therefore, the series converges for all [latex]x[\/latex] such that [latex]|x| &lt; R[\/latex], and the series falls into case iii.<\/p>\r\n<p id=\"fs-id1170572408193\">[latex]_\\blacksquare[\/latex]<\/p>\r\n\r\n<\/section>\r\n<p class=\"whitespace-normal break-words\">If a series [latex]\\displaystyle\\sum_{n=0}^{\\infty} c_n (x-a)^n[\/latex] falls into the third case and converges in an interval, then we have the most common and interesting situation. The series converges for all [latex]x[\/latex] such that [latex]|x-a| &lt; R[\/latex] for some [latex]R &gt; 0[\/latex], and diverges for all [latex]x[\/latex] such that [latex]|x-a| &gt; R[\/latex].<\/p>\r\n<p class=\"whitespace-normal break-words\">What happens at the boundary values where [latex]|x-a| = R[\/latex]? The series may converge or diverge \u2014 we need to check these endpoints individually.<\/p>","rendered":"<h2 data-type=\"title\">Convergence of a Power Series<\/h2>\n<p class=\"whitespace-normal break-words\">Now we need to address a crucial question: for which values of [latex]x[\/latex] does a power series actually converge? Unlike regular series with constant terms, power series behavior depends entirely on the value of [latex]x[\/latex]. The series might converge for some [latex]x[\/latex]-values and diverge for others.<\/p>\n<section class=\"textbox tryIt\" aria-label=\"Try It\"><iframe loading=\"lazy\" id=\"ohm311402\" class=\"resizable\" src=\"https:\/\/ohm.lumenlearning.com\/multiembedq.php?id=311402&theme=lumen&iframe_resize_id=ohm311402&source=tnh\" width=\"100%\" height=\"150\"><\/iframe><\/section>\n<p class=\"whitespace-normal break-words\">Every power series has at least one guaranteed convergence point: <strong>its center<\/strong>.<\/p>\n<p class=\"whitespace-pre-wrap break-words\">For a power series [latex]\\displaystyle\\sum_{n=0}^{\\infty} c_n (x-a)^n[\/latex] centered at [latex]x = a[\/latex], when we substitute [latex]x = a[\/latex]:<\/p>\n<p class=\"whitespace-pre-wrap break-words\" style=\"text-align: center;\">[latex]\\displaystyle\\sum_{n=0}^{\\infty} c_n (a-a)^n = c_0 + 0 + 0 + \\cdots = c_0[\/latex]<\/p>\n<p class=\"whitespace-normal break-words\">The series reduces to just the constant term [latex]c_0[\/latex], so it always converges at its center.<\/p>\n<p class=\"whitespace-normal break-words\">Beyond the center, a power series can behave in exactly one of three ways.<\/p>\n<section class=\"textbox keyTakeaway\" aria-label=\"Key Takeaway\">\n<h3>convergence theorem for power series<\/h3>\n<p class=\"whitespace-normal break-words\">For any power series [latex]\\displaystyle\\sum_{n=0}^{\\infty} c_n (x-a)^n[\/latex], exactly one of these occurs:<\/p>\n<ol class=\"[&amp;:not(:last-child)_ul]:pb-1 [&amp;:not(:last-child)_ol]:pb-1 list-decimal space-y-1.5 pl-7\">\n<li class=\"whitespace-normal break-words\"><strong>Converges only at the center:<\/strong> The series converges at [latex]x = a[\/latex] and diverges everywhere else<\/li>\n<li class=\"whitespace-normal break-words\"><strong>Converges everywhere:<\/strong> The series converges for all real numbers [latex]x[\/latex]<\/li>\n<li class=\"whitespace-normal break-words\"><strong>Converges in an interval:<\/strong> There exists [latex]R > 0[\/latex] such that:\n<ul class=\"[&amp;:not(:last-child)_ul]:pb-1 [&amp;:not(:last-child)_ol]:pb-1 list-disc space-y-1.5 pl-7\">\n<li class=\"whitespace-normal break-words\">The series converges when [latex]|x-a| < R[\/latex]<\/li>\n<li class=\"whitespace-normal break-words\">The series diverges when [latex]|x-a| > R[\/latex]<\/li>\n<li class=\"whitespace-normal break-words\">At the boundary points where [latex]|x-a| = R[\/latex], the series may converge or diverge<\/li>\n<\/ul>\n<\/li>\n<\/ol>\n<\/section>\n<section class=\"textbox connectIt\">\n<p style=\"text-align: center;\"><strong>Proof<\/strong><\/p>\n<hr \/>\n<p id=\"fs-id1170572107834\">Suppose that the power series is centered at [latex]a=0[\/latex]. (For a series centered at a value of <em data-effect=\"italics\">a<\/em> other than zero, the result follows by letting [latex]y=x-a[\/latex] and considering the series [latex]{\\displaystyle\\sum _{n=1}^{\\infty}} {c}_{n} {y}^{n}[\/latex].) We must first prove the following fact:<\/p>\n<p id=\"fs-id1170572109983\">If there exists a real number [latex]d\\ne 0[\/latex] such that [latex]\\displaystyle\\sum _{n=0}^{\\infty }{c}_{n}{d}^{n}[\/latex] converges, then the series [latex]\\displaystyle\\sum _{n=0}^{\\infty }{c}_{n}{x}^{n}[\/latex] converges absolutely for all <em data-effect=\"italics\">x<\/em> such that [latex]|x|<|d|[\/latex].<\/p>\n<p id=\"fs-id1170572366747\">Since [latex]\\displaystyle\\sum _{n=0}^{\\infty }{c}_{n}{d}^{n}[\/latex] converges, the <em data-effect=\"italics\">n<\/em>th term [latex]{c}_{n}{d}^{n}\\to 0[\/latex] as [latex]n\\to \\infty[\/latex]. Therefore, there exists an integer <em data-effect=\"italics\">N<\/em> such that [latex]|{c}_{n}{d}^{n}|\\le 1[\/latex] for all [latex]n\\ge N[\/latex]. Writing<\/p>\n<div id=\"fs-id1170571538405\" class=\"unnumbered\" style=\"text-align: center;\" data-type=\"equation\" data-label=\"\">[latex]|{c}_{n}{x}^{n}|=|{c}_{n}{d}^{n}|{|\\frac{x}{d}|}^{n}[\/latex],<\/div>\n<p>&nbsp;<\/p>\n<p id=\"fs-id1170571637176\">we conclude that, for all [latex]n\\ge N[\/latex],<\/p>\n<div id=\"fs-id1170572436828\" class=\"unnumbered\" style=\"text-align: center;\" data-type=\"equation\" data-label=\"\">[latex]|{c}_{n}{x}^{n}|\\le {|\\frac{x}{d}|}^{n}[\/latex].<\/div>\n<p>&nbsp;<\/p>\n<p id=\"fs-id1170571611376\">The series<\/p>\n<div id=\"fs-id1170572375703\" class=\"unnumbered\" style=\"text-align: center;\" data-type=\"equation\" data-label=\"\">[latex]\\displaystyle\\sum _{n=N}^{\\infty }{|\\frac{x}{d}|}^{n}[\/latex]<\/div>\n<p>&nbsp;<\/p>\n<p id=\"fs-id1170571653945\">is a geometric series that converges if [latex]|\\frac{x}{d}|<1[\/latex]. Therefore, by the comparison test, we conclude that [latex]\\displaystyle\\sum _{n=N}^{\\infty }{c}_{n}{x}^{n}[\/latex] also converges for [latex]|x|<|d|[\/latex]. Since we can add a finite number of terms to a convergent series, we conclude that [latex]\\displaystyle\\sum _{n=0}^{\\infty }{c}_{n}{x}^{n}[\/latex] converges for [latex]|x|<|d|[\/latex].<\/p>\n<p id=\"fs-id1170571595417\">With this result, we can now prove the theorem. Consider the series<\/p>\n<div id=\"fs-id1170572250450\" class=\"unnumbered\" style=\"text-align: center;\" data-type=\"equation\" data-label=\"\">[latex]\\displaystyle\\sum _{n=0}^{\\infty }{a}_{n}{x}^{n}[\/latex]<\/div>\n<p>&nbsp;<\/p>\n<p id=\"fs-id1170572224801\">and let <em data-effect=\"italics\">S<\/em> be the set of real numbers for which the series converges. Suppose that the set [latex]S=\\left\\{0\\right\\}[\/latex]. Then the series falls under case i. Suppose that the set <em data-effect=\"italics\">S<\/em> is the set of all real numbers. Then the series falls under case ii. Suppose that [latex]S\\ne \\left\\{0\\right\\}[\/latex] and <em data-effect=\"italics\">S<\/em> is not the set of real numbers. Then there exists a real number [latex]x*\\ne 0[\/latex] such that the series does not converge. Thus, the series cannot converge for any <em data-effect=\"italics\">x<\/em> such that [latex]|x|>|x*|[\/latex]. Therefore, the set <em data-effect=\"italics\">S<\/em> must be a bounded set, which means that it must have a smallest upper bound. (This fact follows from the Least Upper Bound Property for the real numbers, which is beyond the scope of this text and is covered in real analysis courses.) Call that smallest upper bound <em data-effect=\"italics\">R<\/em>. Since [latex]S\\ne \\left\\{0\\right\\}[\/latex], the number [latex]R < 0[\/latex]. Therefore, the series converges for all [latex]x[\/latex] such that [latex]|x| < R[\/latex], and the series falls into case iii.<\/p>\n<p id=\"fs-id1170572408193\">[latex]_\\blacksquare[\/latex]<\/p>\n<\/section>\n<p class=\"whitespace-normal break-words\">If a series [latex]\\displaystyle\\sum_{n=0}^{\\infty} c_n (x-a)^n[\/latex] falls into the third case and converges in an interval, then we have the most common and interesting situation. The series converges for all [latex]x[\/latex] such that [latex]|x-a| < R[\/latex] for some [latex]R > 0[\/latex], and diverges for all [latex]x[\/latex] such that [latex]|x-a| > R[\/latex].<\/p>\n<p class=\"whitespace-normal break-words\">What happens at the boundary values where [latex]|x-a| = R[\/latex]? The series may converge or diverge \u2014 we need to check these endpoints individually.<\/p>\n","protected":false},"author":15,"menu_order":6,"template":"","meta":{"_candela_citation":"[]","pb_show_title":"on","pb_short_title":"","pb_subtitle":"","pb_authors":[],"pb_section_license":""},"chapter-type":[],"contributor":[],"license":[],"part":673,"module-header":"- Select Header -","content_attributions":[],"internal_book_links":[],"video_content":null,"cc_video_embed_content":{"cc_scripts":"","media_targets":[]},"try_it_collection":null,"_links":{"self":[{"href":"https:\/\/content.one.lumenlearning.com\/calculus2\/wp-json\/pressbooks\/v2\/chapters\/922"}],"collection":[{"href":"https:\/\/content.one.lumenlearning.com\/calculus2\/wp-json\/pressbooks\/v2\/chapters"}],"about":[{"href":"https:\/\/content.one.lumenlearning.com\/calculus2\/wp-json\/wp\/v2\/types\/chapter"}],"author":[{"embeddable":true,"href":"https:\/\/content.one.lumenlearning.com\/calculus2\/wp-json\/wp\/v2\/users\/15"}],"version-history":[{"count":6,"href":"https:\/\/content.one.lumenlearning.com\/calculus2\/wp-json\/pressbooks\/v2\/chapters\/922\/revisions"}],"predecessor-version":[{"id":2324,"href":"https:\/\/content.one.lumenlearning.com\/calculus2\/wp-json\/pressbooks\/v2\/chapters\/922\/revisions\/2324"}],"part":[{"href":"https:\/\/content.one.lumenlearning.com\/calculus2\/wp-json\/pressbooks\/v2\/parts\/673"}],"metadata":[{"href":"https:\/\/content.one.lumenlearning.com\/calculus2\/wp-json\/pressbooks\/v2\/chapters\/922\/metadata\/"}],"wp:attachment":[{"href":"https:\/\/content.one.lumenlearning.com\/calculus2\/wp-json\/wp\/v2\/media?parent=922"}],"wp:term":[{"taxonomy":"chapter-type","embeddable":true,"href":"https:\/\/content.one.lumenlearning.com\/calculus2\/wp-json\/pressbooks\/v2\/chapter-type?post=922"},{"taxonomy":"contributor","embeddable":true,"href":"https:\/\/content.one.lumenlearning.com\/calculus2\/wp-json\/wp\/v2\/contributor?post=922"},{"taxonomy":"license","embeddable":true,"href":"https:\/\/content.one.lumenlearning.com\/calculus2\/wp-json\/wp\/v2\/license?post=922"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}