W. EUGENE DAVIS, Circuit Judge:

Plaintiffs-Appellees, a putative class of plaintiffs, seek to recover damages from Defendants-Appellants for securities fraud under § 10(b) of the Securities Exchange Act of 1934. The district court concluded that Defendants-Appellants were not entitled to use evidence of no market price impact to rebut the fraud-on-the-market presumption of reliance at class certification. We AFFIRM.

I.

This litigation arises out of alleged misrepresentations by the Halliburton Company and its CEO, President, and Chairman of the Board, David Lesar (collectively "Halliburton"). The Plaintiffs-Appellees, represented by the Erica P. John Fund, Inc. ("the Fund"), are a putative class of shareholders who allege that they suffered material losses as a result of these fraudulent misrepresentations between June 3, 1999, and December 7, 2001. Over this period of time, the Fund contends that Halliburton made misrepresentations concerning three primary aspects of its operations: (1) it understated its projected liability for asbestos claims, (2) it overstated its revenues by including billings whose collections were unlikely, and (3) it exaggerated the cost savings and efficiencies Halliburton would derive from its 1998 merger with Dresser Industries. Plaintiffs allege that these misrepresentations temporarily and artificially inflated the price of Halliburton stock; when the truth was subsequently revealed, the stock price fell, causing damages to those who purchased the stock in the relevant timeframe.

In September 2007, the Fund moved to certify a class of all persons who purchased Halliburton's common stock during the class period. The district court first determined that the Fund had satisfied the Fed.R.Civ.P. 23(a) threshold class certification requirements of numerosity, commonality, typicality, and adequacy of representation. Turning to Rule 23(b)(3)'s predominance requirement, the district court conducted a limited inquiry into the plaintiffs' cause of action to determine whether common questions of law and fact predominated over questions affecting only individual plaintiffs. The court observed that "the Fifth Circuit has placed an extremely high burden on plaintiffs seeking class certification in a securities fraud case." Specifically, the court pointed to Fifth Circuit precedent requiring securities fraud plaintiffs to make a showing of loss causation before obtaining certification. The district court then found that plaintiffs had not established loss causation and declined to certify the class. See Archdiocese of Milwaukee Supporting Fund, Inc. v. Halliburton Co., No. 3:02-CV-1152-M, 2008 WL 4791492 (N.D.Tex. Nov. 14, 2008) (unpublished).

On appeal, a panel of this court affirmed the district court's denial of class certification based on its conclusion that the Fund had "failed to meet this court's requirements for proving loss causation at the class certification stage." See Archdiocese of Milwaukee Supporting Fund, Inc. v. Halliburton Co., 597 F.3d 330, 344 (5th Cir.2010) ("AMS Fund"). The Fund filed a petition for a writ of certiorari, which the Supreme Court granted. In Erica P. John Fund, Inc. v. Halliburton Co., ___ U.S. ___, 131 S.Ct. 2179, 2184, 180 L.Ed.2d 24 (2011) ("EPJ Fund"),1 a unanimous Supreme Court reversed the judgment of the Fifth Circuit, finding that this court "erred by requiring proof of loss causation for class certification." The Court then remanded the case back to this court, stating, "To the extent Halliburton has preserved any further arguments against class certification, they may be addressed in the first instance by the Court of Appeals on remand." Id. at 2187.

On remand from the Supreme Court, this court remanded the case to the district court for further proceedings. Halliburton argued to the district court that the class should still not be certified because Halliburton's class certification evidence revealed that its alleged fraud did not affect the market price of the stock; that is, its alleged misrepresentation did not cause "price impact" or "price distortion." The district court declined to consider Halliburton's evidence on the issue, finding that price impact evidence did not bear on the critical inquiry of whether common issues predominated under Rule 23(b)(3). Based on its finding that common issues predominated and that the other Rule 23 class prerequisites were satisfied, the district court certified the class. Halliburton now appeals.

II.

We review the district court's class certification decision for abuse of discretion. See Benavides v. Chicago Title Ins. Co., 636 F.3d 699, 701 (5th Cir.2011). "Because, however, a court by definition abuses its discretion when it applies an incorrect legal standard, we review such errors de novo." Id. While the district court has substantial discretion to grant or deny certification, it "must conduct a rigorous analysis of the rule 23 prerequisites before certifying a class." Id.

III.

A.

Halliburton's primary argument on appeal is that the district court erred by not permitting Halliburton to challenge class certification with evidence that the alleged misrepresentations did not impact the price of the stock (i.e., there was no price impact).

1.

The pivotal question in this case is whether a defendant should be permitted to show the absence of price impact at the class certification stage of the proceedings to establish that common issues among class members do not predominate and that class certification is inappropriate.

A potential class of securities fraud plaintiffs, like any other group seeking class certification, must satisfy the requirements of Fed.R.Civ.P. 23 in order to be certified. Rule 23 provides that a class action may be maintained if the conditions of 23(a) and (b) are met. To satisfy the criteria set forth in Rule 23(a), a plaintiff must demonstrate numerosity, commonality, typicality, and adequacy of representation. In this case, the plaintiff must also show that common questions predominate, as provided in Rule 23(b). The parties agree that Rule 23(a)'s requirements have been met, so that the only element at issue is whether common questions predominate. Thus, if "questions of law or fact common to class members predominate over any questions affecting only individual members," then a class of purchasers of Halliburton common stock from June 3, 1999-December 7, 2001 should be certified. FED.R.CIV.P. 23(b)(3).

The private securities fraud action is based upon federal securities statutes and their implementing regulations. Section 10(b) of the Securities Exchange Act of 1934 forbids the use of any "deceptive device," in "connection with the purchase or sale of any security." 15 U.S.C. § 78j(b) (2006). The Securities and Exchange Commission's corresponding Rule 10b-5 forbids, among other things, the making of any "untrue statement of a material fact" "in connection with the purchase or sale of any security." 17 CFR § 240.10b-5 (2004). From these provisions, courts have derived a private securities fraud cause of action; to succeed at trial or summary judgment, a plaintiff is required to establish the 10b-5 action's elements: (1) a material misrepresentation, (2) scienter (deceptive intent), (3) a connection with the purchase or sale of a security, (4) reliance, (5) economic loss, and (6) loss causation. Dura Pharms., Inc. v. Broudo, 544 U.S. 336, 341-42, 125 S.Ct. 1627, 161 L.Ed.2d 577 (2005).

The burden of establishing all the requirements of class certification likewise falls on the party seeking certification, here the Fund. See Wal-Mart Stores, Inc. v. Dukes, ___ U.S. ___, 131 S.Ct. 2541, 2551, 180 L.Ed.2d 374 (2011). Establishing common question predominance as a prerequisite to class certification is different from the burden of proving 10b-5 fraud on the merits, although the inquiries may overlap. As the Supreme Court has stated, "[w]hether common questions of law or fact predominate in a securities fraud action often turns on the element of reliance." EPJ Fund, 131 S.Ct. at 2184. Because common question predominance in this context hinges on reliance, there has been considerable debate concerning what evidence relating to reliance is required or allowed at class certification. The Supreme Court has issued several decisions touching on this issue, which we now turn to for guidance.

2.

The Court in Basic Inc. v. Levinson first considered the difficulty inherent in establishing proof of class-wide reliance. 485 U.S. 224, 108 S.Ct. 978, 99 L.Ed.2d 194 (1988). As the Basic Court stated, "Reliance provides the requisite causal connection between a defendant's misrepresentation and a plaintiff's injury." Id. at 243, 108 S.Ct. 978. "The traditional (and most direct) way a plaintiff can demonstrate reliance is by showing that he was aware of a company's statement and engaged in a relevant transaction — e.g., purchasing common stock — based on that specific misrepresentation." EPJ Fund, 131 S.Ct. at 2185. However, if this were the only way to prove reliance, it "would place an unnecessarily unrealistic evidentiary burden on the [securities fraud] plaintiff who has traded on an impersonal market." Basic, 485 U.S. at 245, 108 S.Ct. 978. Without some legal accommodation, the element of reliance would become a "barrier to class certification, since each of the individual investors would have to prove reliance on the alleged misrepresentation." Dukes, 131 S.Ct. at 2552 n. 6.

As a result, the Basic Court adopted a legal presumption previously accepted by several circuits: the "fraud-on-the-market presumption" of reliance. The Supreme Court explained that because the market price of a security in an efficient market will immediately incorporate any material, public representation, a purchaser who buys a security at the market price will be presumed to have relied upon the representation:

Basic, 485 U.S. at 244, 108 S.Ct. 978 (quoting In re LTV Sec. Litig., 88 F.R.D. 134, 143 (N.D.Tex.1980)).

To invoke the fraud-on-the-market presumption, a plaintiff must establish the prerequisites necessary for market price incorporation of information: (1) misrepresentation publicity, (2) misrepresentation materiality, (3) market efficiency, and (4) that the plaintiff traded the shares between the time the misrepresentations were made and the time the truth was revealed. Id. at 248 n. 27, 108 S.Ct. 978.

Basic also established that the defendant is entitled to rebut the fraud-on-the-market presumption by demonstrating certain facts that undermine its basic assumptions. "Any showing that severs the link between the misrepresentation and either the price received (or paid) by the plaintiff, or his decision to trade at a fair market price, will be sufficient to rebut the presumption of reliance." Id. at 248, 108 S.Ct. 978. For example, there would be no reliance if a misrepresentation was ignored because the truth was well-known to the market, or there would be no reliance by those who purchased the security after the truth had already entered the market and dissipated the effects of the fraud. Id. at 248-49, 108 S.Ct. 978. If the defendant could make such a showing, the market price could not be said to have incorporated — or relied upon — the misrepresentation, and "the basis for finding that the fraud had been transmitted through market price would be gone." Id. Similarly, if a defendant could demonstrate that a plaintiff knew about the misrepresentation but decided to purchase the stock anyway, the plaintiff would not have relied upon the integrity of the market price. Id. at 249, 108 S.Ct. 978. Though making clear that the fraud-on-the-market presumption could be rebutted, the Court in Basic did not decide the extent to which the presumption could be rebutted at class certification.

After Basic, the circuits eventually began to apply the fraud-on-the-market presumption inconsistently. As a result, the Supreme Court stepped in to clarify the law again in EPJ Fund v. Halliburton, ___ U.S. ___, 131 S.Ct. 2179, 180 L.Ed.2d 24 (2011), when the Court considered another issue in its review of an earlier decision by this court in the instant case. The specific question before the Supreme Court in EPJ Fund was whether proof of loss causation was required to invoke the fraud-on-the-market presumption of reliance at the class certification stage. In AMS Fund,2 this court found that in order for plaintiffs to invoke the fraud-on-the-market presumption of reliance and obtain class certification, they must establish proof of loss causation; this requires proof that the stock price declined after the false statement is corrected and the truth is revealed.3 Because the Fund had not established loss causation, we affirmed the district court's denial of class certification. See AMS Fund, 597 F.3d at 339-43.

After granting certiorari, the Supreme Court acted swiftly to correct this court's error, rejecting our position that a putative class had to establish loss causation as a predicate to invoking the fraud-on-the-market presumption of reliance at class certification: "The Court of Appeals' requirement is not justified by Basic or its logic.... Loss causation addresses a matter different from whether an investor relied on a misrepresentation, presumptively or otherwise, when buying or selling a stock." EPJ Fund, 131 S.Ct. at 2186. The Court held that proof of loss causation, or a decline in stock value after revelation of the truth, is a conceptually distinct inquiry and is not necessary to establish reliance. Id. at 2185-86. The EPJ Fund Court declined, however, to address Halliburton's assertion that it was entitled to rebut the presumption at class certification for other reasons. The Court stated: "[T]he Court of Appeals erred by requiring EPJ to prove loss causation at the certification stage, ... and [we] do not, address any other question about [the fraud-on-the-market] presumption, or how and when it may be rebutted." Id. at 2187.

The Supreme Court had to intervene yet a third time as courts struggled with the question of which issues a securities fraud plaintiff had to prove in order to invoke the fraud-on-the-market presumption at class certification, and the corollary question of whether the defendant could present rebuttal evidence on the issues. See Amgen Inc. v. Conn. Ret. Plans and Trust Funds, ___ U.S. ___, 133 S.Ct. 1184, 185 L.Ed.2d 308 (2013). Although it was clear that plaintiffs must eventually establish the materiality of a misrepresentation on the merits to invoke the fraud-on-the-market presumption, the precise questions before the Amgen Court were whether the putative class must establish materiality at class certification to invoke the presumption; and relatedly, whether a defendant is entitled to rebut the presumption at class certification with proof of immateriality. Id. at 1194.

The Amgen Court began by emphasizing that the central issue in resolving this question was whether proof of materiality was required to satisfy Rule 23(b)(3)'s requirement of common question predominance. Id. at 1195. Thus, the twin issues of what a plaintiff must prove and what a defendant may rebut at class certification are resolved by the same inquiry: "[T]he pivotal inquiry is whether proof of materiality is needed to ensure that the questions of law or fact common to the class will `predominate over any questions affecting only individual members' as the litigation progresses." Id. (quoting FED.R.CIV.P. 23(b)(3)).4 Thus, the focus of the 23(b)(3) class certification inquiry — predominance — is not whether the plaintiffs will fail or succeed, but whether they will fail or succeed together. Id. at 1197. Although the other prerequisites to invoke the fraud-on-the-market presumption — purchase timing, publicity, and market efficiency — must be established at class certification, the same is not true of materiality. Id. at 1198-99. The Court held, "While [a putative class] certainly must prove materiality to prevail on the merits, we hold that such proof is not a prerequisite to class certification." Id. at 1191. The Amgen Court thus made clear that what is required for a plaintiff to "invoke" the fraud-on-the-market presumption on the merits is not necessarily what is required for the plaintiff to benefit from the presumption at class certification.

The Court based its determination that materiality need not be established at class certification on the answers to two crucial questions: (1) whether the question of materiality was an objective inquiry that could "be proved through evidence common to the class;" and (2) whether there was a risk that a failure of proof on the question of materiality would "result in individual questions predominating." Id. at 1195-96. The Court determined that because materiality is established by evidence common to all plaintiffs, and because a failure to prove materiality will cause all plaintiffs' individual claims to fail, materiality evidence was not relevant at class certification. Id. at 1197. In summary, the Court stated: "As to materiality, therefore, the class is entirely cohesive: It will prevail or fail in unison. In no event will the individual circumstances of particular class members bear on the inquiry." Id. at 1191. It is this characteristic of materiality which distinguishes it from the other fraud-on-the-market presumption's prerequisites. Id. at 1198-99. A plaintiff can fail to establish publicity, market efficiency, or trade timing, and therefore lose the class-wide presumption of reliance, but still establish individual reliance and prove fraud. Id. Thus, only those issues which bear directly on the pivotal inquiry of common question predominance and the propriety of class resolution should be addressed at class certification. Id.

3.

We thus know from the above Supreme Court cases that in order for a 10b-5 plaintiff to invoke the fraud-on-the-market presumption of reliance on the merits, a plaintiff must establish (1) trade timing, (2) market efficiency, (3) publicity, and (4) materiality. See id. at 1192-93. However, the fraud-on-the-market elements that should be addressed at class certification are limited to those matters which bear on common question predominance and the propriety of class resolution: trade timing, market efficiency, and publicity (but not materiality).5 Id. at 1194-99.

Halliburton frames the question before us as whether price impact is an issue which a defendant may address at class certification to rebut the fraud-on-the-market presumption that the stock price was affected by a misrepresentation. According to EPJ Fund, "`Price impact' simply refers to the effect of a misrepresentation on a stock price."6 131 S.Ct. at 2187. It is neither an element of 10b-5 fraud nor an element of the fraud-on-the-market theory. In fact, price impact evidence does not fit neatly into any one fraud issue, but is probative of materiality, statement publicity, and market efficiency, all of which are relevant in establishing the presumption of fraud-on-the-market reliance.7 According to Amgen, however, only some of these matters may be considered at class certification. 133 S.Ct. at 1197-99. For example, price impact evidence relating to materiality may not be considered at class certification (because materiality does not bear on Rule 23(b)(3) common question predominance), but price impact evidence relating to market efficiency or statement publicity could be considered. See id. We must therefore determine at what issue Halliburton's price impact evidence is directed.

In this case, Halliburton contends that its price impact evidence is not intended to rebut materiality, market efficiency, or statement publicity — the issues which the Supreme Court has specifically addressed. Rather, Halliburton contends that its price impact evidence is intended only to generally rebut the fraud-on-the-market presumption of reliance without necessarily attacking one of the presumption's individual elements. More specifically, Halliburton argues that despite the proof offered in support of invoking the fraud-on-the-market presumption of reliance, its evidence shows that the price did not actually transfer the effects of the alleged fraud to a stock purchaser.

As the Basic Court stated, "Any showing that severs the link between the alleged misrepresentation and either the price received (or paid) by the plaintiff ... will be sufficient to rebut the presumption of reliance." 485 U.S. at 248, 108 S.Ct. 978. We agree with Halliburton that in the absence of price impact, "the basis for finding that fraud has been transmitted through market price would be gone." See id. Accordingly, Halliburton's price impact evidence potentially demonstrates that despite the presence of the necessary conditions for market price incorporation of fraudulent information (fraud-on-the-market reliance), no such incorporation occurred in fact. Thus, Halliburton's price impact evidence could be used at the trial on the merits to refute the presumption of reliance.8 See id. at 248-49, 108 S.Ct. 978. The Amgen Court did not discuss whether evidence offered for this purpose could be considered at class certification, but it did set forth the proper analytical framework so that we may resolve the question.

As the Amgen court made clear, the "pivotal inquiry" when determining whether to consider a matter at class certification is whether resolution of the matter "is needed to ensure that the questions of law or fact common to the class will `predominate over any questions affecting only individual members' as the litigation progresses." Amgen, 133 S.Ct. at 1195. Amgen held that materiality and the rebuttal of materiality were not issues to be considered at class certification because this proof depended on evidence common to all class members; moreover, the failure of plaintiffs to prevail on the issue of materiality would not cause individual issues to "overwhelm questions common to the class, for the class members' claims will have failed on their merits, thus bringing the litigation to a close." Id. at 1204.

Turning to the instant case, the first question we ask is whether price impact evidence is common to the class. Because price impact is simply a measure of the effect of a misrepresentation on a security's price, it is undoubtedly an objective inquiry. See EPJ Fund, 131 S.Ct. at 2187. As the record evidence in this case demonstrates, price impact is ordinarily established by expert evaluation of a stock's market price following a specific event and it inherently applies to everyone in the class. The first Amgen consideration therefore suggests that price impact fraud-on-the-market rebuttal evidence should not be addressed at class certification.

The second inquiry suggested by Amgen is whether there is any risk that a later failure of proof on the common question of price impact will result in individual questions predominating. See Amgen, 133 S.Ct. at 1196. In Amgen, the Court found that a failure to establish materiality could not result in the continuation of any individual claims, because immateriality would be fatal to all plaintiffs' claims. Id. Because materiality is an element of every fraud claim, immateriality absolutely destroys both class and individual causes of action. The absence of materiality "end[s] the case for one and for all." Id.

Turning to the instant case, we must determine whether the failure to prove price impact will necessarily cause all plaintiffs' claims to fall together. In other words, if Halliburton successfully rebuts the fraud-on-the-market presumption with evidence of no price impact, could individual plaintiffs still proceed with their fraud claims? Halliburton contends that a failure on the part of the plaintiffs to prove price impact will not cause all claims to fail, because unlike materiality, price impact is not a required element of fraud. Thus, Halliburton argues, a plaintiff class which fails to show price impact would only lose the class-wide presumption of reliance, leaving individual plaintiffs with viable fraud claims.

We disagree. Although the 10b-5 fraud action does not expressly require proof of price impact as an element of the claim, a plaintiff must nevertheless prevail on this fact in order to establish another element on which the plaintiff does bear the burden of proof: loss causation. As the Court in EPJ Fund stated, "`Price impact' simply refers to the effect of a misrepresentation on a stock price." 131 S.Ct. at 2187. Price impact can be shown either by an increase in price following a fraudulent public statement or a decrease in price following a revelation of the fraud. To successfully prove a lack of price impact, Halliburton would thus be required to demonstrate both that the stock price did not increase when the misrepresentation was announced, and that the price did not decrease when the truth was revealed. If Halliburton were to successfully show that the price did not drop when the truth was revealed, then no plaintiff could establish loss causation. See id. at 2185.9 In other words, because a showing of negative price impact is required to establish loss causation, plaintiffs who cannot establish price impact cannot establish loss causation. Thus, if Halliburton were to successfully rebut the fraud-on-the-market presumption by proving no price impact, the claims of all individual plaintiffs would fail because they could not establish an essential element of the fraud action. In the words of the Amgen Court, "[T]he class members' claims will have failed on their merits, thus bringing the litigation to a close." See Amgen, 133 S.Ct. at 1204. Thus, the second Amgen consideration also leads to the conclusion that price impact fraud-on-the-market rebuttal evidence should not be addressed at class certification.10

Halliburton's only other argument is its policy-based contention that not considering evidence of price impact at class certification will enhance the "in terrorem power of certification," and allow plaintiffs to extort non-meritorious settlements from corporate defendants. See Oscar, 487 F.3d at 267. The Supreme Court rejected an identical argument in Amgen, pointing out that "Congress has homed in on the precise policy concerns raised" by this argument, but has selected different remedies. See Amgen, 133 S.Ct. at 1201.

The Amgen Court's analysis leads to the conclusion that price impact fraud-on-the-market rebuttal evidence should not be considered at class certification.11 Proof of price impact is based upon common evidence, and later proof of no price impact will not result in the possibility of individual claims continuing. Accordingly, Halliburton's price impact evidence does not bear on the question of common question predominance, and is thus appropriately considered only on the merits after the class has been certified.12

B.

The Fund also argues that because Halliburton waived the argument it now presents by failing to initially raise it before the district court, the argument has not been preserved and we may not consider it.13 This waiver issue is not subject to our discretion, the Fund contends, because the Supreme Court expressly limited our consideration of issues on appeal to those which Halliburton had "preserved."

When the Supreme Court remanded the instant case back to this court, it instructed us to consider Halliburton's arguments against class certification "[t]o the extent Halliburton has preserved" them. See EPJ Fund, 131 S.Ct. at 2187. As the Supreme Court has consistently noted, "an inferior court has no power or authority to deviate from the mandate issued by an appellate court." Briggs v. Penn. R.R. Co., 334 U.S. 304, 306, 68 S.Ct. 1039, 92 L.Ed. 1403 (1948). We thus may consider Halliburton's fraud-on-the-market rebuttal argument only to the extent it has been preserved.

Ordinarily, "arguments not raised before the district court are waived and cannot be raised for the first time on appeal." Martco Ltd. P'ship v. Wellons, Inc., 588 F.3d 864, 877 (5th Cir.2009). The Fund insists that because Halliburton never attempted to rebut fraud-on-the-market reliance before the district court, the Supreme Court's decree prevents this court from considering the argument now. The Fund contends that Halliburton attempted to use its price impact evidence to rebut the element of loss causation but did not challenge the element of reliance; having lost on the loss causation issue before the EPJ Fund Court, Halliburton should not now be able to challenge the Fund's fraud-on-the-market reliance.

However, the Fund's argument ignores both the substance of Halliburton's pleadings and the state of the fraud-on-the-market theory as interpreted by this circuit before EPJ Fund. Although Halliburton's reliance-rebuttal argument has technically been available since Basic, this court had since applied a significant judicial gloss to Basic. In fact, our cases required defendants to use evidence of no price impact to undermine the fraud-on-the-market theory in another way — by rebutting loss causation.14 Halliburton did not direct its evidence of no price impact at fraud-on-the-market reliance in an effort to comply with our case law: under our pre-EPJ Fund framework, evidence directed at loss causation was by definition directed at fraud-on-the-market reliance. The Supreme Court has since corrected our fraud-on-the-market framework and established that loss causation is not relevant to the fraud-on-the-market presumption of reliance. EPJ Fund, 131 S.Ct. at 2185-86. However, we decline to penalize Halliburton for framing its evidence in the manner we instructed. It is well-settled that when the law changes in unanticipated ways during an appeal, parties are generally given an opportunity to apply the new law and present arguments relevant to the new standard. See Deffenbaugh-Williams v. Wal-Mart Stores, Inc., 188 F.3d 278, 282 (5th Cir.1999).

IV.

For the reasons stated above, the judgment of the district court is AFFIRMED.