This appeal and cross-appeal from the United States Court of International Trade concerns an administrative review of an antidumping order covering extruded rubber thread from Malaysia. The United States appeals from the portion of the judgment holding that the Department of Commerce did not have the authority to conduct a duty absorption inquiry in this case. In light of a recent decision of this court, the United States now agrees with plaintiffs that we should sustain the judgment with respect to that issue; accordingly, we affirm the judgment of the Court of International Trade in that respect. Plaintiffs Heveafil Sdn. Bhd. and Filmax Sdn. Bhd. (collectively, “Heveafil”) cross-appeal from the portion of the judgment sustaining the assignment of an antidumping duty margin based on the statutorily prescribed standards of “facts otherwise available” and “adverse inferences.” We affirm the use of facts otherwise available and adverse inferences as a basis for setting the dumping margin. However, we reverse and remand with respect to the specific dumping rate assigned, because the source of the corroboration of the dumping rate imposed on Heveafil has recently been invalidated and is no longer appropriate as a basis from which to calculate the proper dumping rate in this case.
I
On October 7, 1992, the Department of Commerce published an antidumping duty order addressing extruded rubber thread from Malaysia. Heveafil Sdn. Bhd. and Filmax Sdn. Bhd., affiliated producers of *845extruded rubber thread in Malaysia, are subject to a single dumping margin. In 1996, Commerce began an annual administrative review-the fourth review under the order-at the request of Heveafil, which sought to have its dumping rate reduced. Following completion of the review, Commerce published its final determination in 1998. Extruded Rubber Thread From Malaysia; Final Results of Antidumping Duty Administrative Review, 63 Fed. Reg. 12,752 (Mar. 16,1998).
During the review, Commerce determined that it was necessary to verify the sales and cost information Heveafil had provided. Commerce was able to verify Heveafil’s sales information, but when Heveafil proffered a copy of a bill of materials (“BOM”) database on a computer diskette as a means for verifying its cost information, Commerce refused to accept the diskette as a basis for verification. According to Commerce, Heveafil stated that the BOM database was the primary source of the costs reflected in its questionnaire response, but that the 1996 Budgeting Report also reflected the reported costs. Heveafil informed Commerce that the original BOM database for the period of review was no longer on its mainframe computer system and that the company had no hard copy of the database. Commerce stated in its verification report that
[bjecause hard copies of the underlying source documents had been destroyed (and, consequently, because the company was unable to show that the data had not been manipulated), we did not accept the worksheet offered by company officials. Rather, we informed the company that it needed to demonstrate that its response was complete and accurate using documents generated in the ordinary course of business during the POR and located at the verification site.
Commerce reported that it had attempted to verify the information using a portion of the 1996 Budgeting Report and other source documents, but that its efforts had proved unsuccessful. In its verification report, Commerce focused on two problems it had encountered during verification: the destruction of essential source documents and inadequate preparation by Heveafil, including the Heveafil employees’ lack of familiarity with the accounting system and Heveafil’s failure to copy requested documents. Commerce concluded that Heveafil had failed verification and that, pursuant to 19 U.S.C. § 1677e(a), Commerce would use “facts otherwise available” to formulate Heveafil’s antidumping duty margin.
In its final determination, Commerce also found that Heveafil had failed to cooperate to the best of its ability during the administrative review and that “adverse inferences” should be drawn in selecting among the facts otherwise available, pursuant to 19 U.S.C. § 1677e(b). Commerce made that determination based on the fact that it was “unable to verify the information submitted by Heveafil in this period of review (POR) and because the company failed to adequately prepare and provide information during the verification” despite its experience with prior reviews. 63 Fed. Reg. at 12,753, 12,762. While Commerce considered Heveafil’s explanation that it had experienced significant employee turnover, Commerce determined that it could not conclude “that the company as a whole was so inexperienced as to be unaware of the necessity of retaining key source documents for verification purposes.” Id. at 12,762. Commerce also noted that it had made repeated and unsuccessful attempts to verify the reported data using various sources and that Heveafil had failed to produce the 1996 Budgeting Report at the beginning of the verification process, even though Heveafil’s questionnaire response indicated that the *846Budgeting Report was a basis for the reported cost information. Id. Commerce referred back to its verification report for further discussion of Heveafil’s failure to cooperate; that report addressed in greater detail Heveafil’s lack of preparation for the verification. Id.
The Commerce officials who conducted the review recommended a dumping rate of 10.68 percent — the highest margin calculated for Heveafil in prior reviews. In its final order, however, Commerce assigned a dumping rate of 54.13 percent— the highest margin ever assigned to any producer under the antidumping duty order. See 63 Fed. Reg. at 12,753. In addition, Commerce found that the producers were absorbing antidumping duties under 19 U.S.C. § 1675(a)(4). See id. at 12,757.
Heveafil appealed Commerce’s final determination to the Court of International Trade. That court concluded that Commerce had acted within its discretion when it assigned a dumping margin based on facts otherwise available. It also concluded that Commerce’s use of adverse inferences was lawful and supported by substantial evidence, and that Commerce had adequately corroborated the adverse facts available margin that it assigned to Heveafil under 19 U.S.C. § 1677e(c). The court noted the broad discretion Congress gave to Commerce to devise verification procedures and select methods of verification.
With respect to the particulars of the verification process, the court agreed with Commerce that the copy of the BOM database on the computer diskette was not generated in the ordinary course of business. The court explained that “[a]t best, it was an unauthenticated duplicate of a database which may have been generated within the ordinary course of business” and that Heveafil had presented “no evidence demonstrating that it could not have maintained the BOM in its original state pending verification.” With respect to the use of adverse inferences, the court noted that Heveafil had undergone verification in the past and that it had deleted the BOM data from its mainframe computer system even though it had been informed only six months earlier that it should retain source documents for purposes of verification review. Based on those findings, the court sustained Commerce’s use of adverse inferences in calculating the antidumping duty margin. With respect to the issue of duty absorption, however, the court concluded that Commerce lacked statutory authority to conduct the duty absorption inquiry because the relevant antidumping duty order was issued before January 1, 1995.
In its appeal to this court, the United States challenged the ruling by the Court of International Trade that Commerce’s duty absorption inquiry was not in accordance with law. While the government’s appeal was pending, however, this court issued an opinion in FAG Italia S.p.A. v. United States, 291 F.3d 806 (Fed.Cir. 2002), which affirmed a ruling by the Court of International Trade that Commerce did not have the statutory authority to conduct an inquiry in the second or fourth years after publication of a pre1995 antidumping duty order. In light of that decision, the government now agrees that we should sustain the judgment of the Court of International Trade with respect to the duty absorption inquiry issue.
On cross-appeal, Heveafil argues that Commerce erred in assigning a dumping rate based on facts otherwise available, that Commerce had no grounds for using adverse inferences, and that the dumping rate was excessive. The government responds that Commerce may lawfully use facts otherwise available and adverse inferences to formulate the margin, but it *847acknowledges that even if the use of facts otherwise available and adverse inferences is upheld, this case should be remanded for Commerce to assign a different anti-dumping duty margin, because the rate used by the agency as corroboration for the assigned dumping rate has recently been invalidated.
II
The Department of Commerce has broad discretion in executing the anti-dumping laws. Smith-Corona Group v. United States, 713 F.2d 1568, 1571 (Fed. Cir.1983). This court has repeatedly held that Commerce’s special expertise makes it the “master” of the antidumping law, entitling its decisions to great deference from the courts. F.LLI. De Cecco Di Filippo Fara S. Martino S.p.A. v. United States, 216 F.3d 1027, 1032 (Fed.Cir.2000); Melamine Chems., Inc. v. United States, 732 F.2d 924, 929 (Fed.Cir.1984) (“Congress has afforded Commerce considerable latitude and discretion in implementing the antidumping duty laws”). We hold that the factual findings that led Commerce to invoke the facts otherwise available method of calculating Heveafil’s dumping duty, see 19 U.S.C. § 1677e(a)(2), and to draw adverse inferences against Heveafil in selecting from among the facts otherwise available, see id. § 1677e(b), are supported by substantial evidence and that Commerce did not commit legal error or otherwise abuse its discretion in applying those statutory standards in this case.
A
Heveafil first argues that Commerce erred when it concluded that Heveafil had failed verification of cost information and therefore used facts otherwise available in determining Heveafil’s anti-dumping duty margin. Under the anti-dumping statute, if a respondent fails to provide evidence supporting its reported information and the information provided by the respondent cannot be verified, Commerce can decline to consider that information and can use facts otherwise available to formulate a dumping margin. See 19 U.S.C. § 1677e(a). The applicable Commerce Department regulations do not specify any standards or procedures for verification, and the courts have accorded Commerce broad discretion with regard to the conduct and extent of such investigations. See Micron Tech, Inc. v. United States, 117 F.3d 1386, 1394-95 (Fed.Cir. 1997); Torrington Co. v. United States, 68 F.3d 1347, 1351 (Fed.Cir.1995). “By requiring that Commerce report, on a case-by-case basis, the methods and procedures used to verify submitted information, Congress has implicitly delegated to Commerce the latitude to derive verification procedures ad hoc.” Micron, 117 F.3d at 1396. Accordingly, we have stated that “we review verification procedures employed by Commerce in an investigation for abuse of discretion rather than against previously-set standards.” Id.
In this case, the verification dispute centers on Commerce’s refusal to use a diskette that Heveafil proffered as containing a copy of the BOM database to verify Heveafil’s cost information. Heveafil argues that Commerce did not have a legitimate reason for refusing to use the diskette copy of the database, because Heveafil used the version of the BOM database on the company’s mainframe computer in the ordinary course of business and diskette contained an exact duplicate of the database that had been developed on the company’s mainframe computer.
In light of the deference given to Commerce to develop its own verification procedures, we conclude that Commerce acted within its discretion in rejecting the diskette version of the BOM database. Al*848though Heveafil asserts that the copy of the BOM database on the diskette is an unaltered reproduction of the BOM database that was previously stored and used on the company’s mainframe computer, there is no evidence in the record supporting that assertion. While Commerce could have taken Heveafil’s word for the authenticity of the diskette copy of the BOM database, we cannot conclude that Commerce abused its discretion in rejecting the diskette copy when the authenticity of that purported copy was not established. We do not accept Heveafil’s argument that Commerce was required to accept a copy of a database stored on a separate and easily manipulated medium when Heveafil had provided no confirmation of the veracity of that database.
Reasoning by analogy from the Federal Rules of Evidence, Heveafil contends that Commerce was required to accept the diskette, either as an original under Rule 1001(3) or a duplicate presumed to be accurate under Rules 1001(4) and 1003. Regardless of whether an item such as the diskette database would be considered an original or a duplicate under those rules, however, it would still have to be authenticated before being admissible as evidence. Under Rules 901 and 902 of the Federal Rules of Evidence, a proper foundation for the reliability and trustworthiness of an item such as the diskette database would be required for authentication. In this case, the contents of the diskette have not been shown to reflect the BOM data accurately. Heveafil contends that it could have demonstrated that the diskette data was accurate; that assertion, however, is not supported by anything in the record. For example, at the time it proffered the diskette, Heveafil did not produce evidence explaining how the copy was made, such as an affidavit by an employee with pertinent knowledge verifying the accuracy of the database. Thus, even if the Federal Rules of Evidence applied to this administrative antidumping proceeding, they would not compel Commerce to accept the diskette database.
Commerce attempted to proceed with the verification using other means, such as consulting the 1996 Budgeting Report, but Commerce was unable to do so as a result of Heveafil’s inability to produce a complete copy of that report during the verification process. Moreover, Commerce reported that when its verification officials attempted to reconcile the costs shown in the portion of the 1996 Budgeting Report that was produced at verification, they were unable to do so in a number of instances, and they were unable to reconcile the costs for the products missing from the Budgeting Report to Heveafil’s inventory records.
Because Commerce lacked an authenticated copy of the BOM database or any other source document that it could use to verify the company’s cost information, we conclude that substantial evidence supports Commerce’s determination that Heveafil failed verification. Commerce legitimately concluded that Heveafil did not submit cost information that Commerce was able to verify successfully. It was therefore lawful under 19 U.S.C. § 1677e(a) for Commerce to use facts otherwise available to determine the appropriate dumping rate for the administrative review.
B
Heveafil next argues that Commerce erred in utilizing adverse inferences under 19 U.S.C. § 1677e(b) when selecting among the facts otherwise available and assigning Heveafil a dumping margin. Section 1677e(b) provides that if Commerce “finds that an interested party has failed to cooperate by not acting to the *849best of its ability to comply with a request for information,” Commerce may “use an inference that is adverse to the interests of that party in selecting from among the facts otherwise available.” Heveafíl contends that Commerce did not have grounds for finding that Heveafíl failed to cooperate to the best of its ability, and that Commerce improperly presumed that Heveafil did not cooperate simply because it could not provide the cost verification information. Arguing that it had to update the BOM database on its mainframe computer system to keep the database current and to keep its financial statements accurate, Heveafíl states that its actions were necessary to continue business operations and that its transfer of the pertinent BOM data to a diskette was intended to preserve, rather than destroy, the data for Commerce. In addition, Heveafíl contends that Commerce failed to take into account evidence of the company’s various efforts to cooperate with the verification process.
We conclude that substantial evidence supports Commerce’s express determination that Heveafíl “did not cooperate to the best of its ability in verifying its reported cost data.” The absence of acceptable source documents was not the only problem encountered by Commerce. According to Commerce, Heveafil’s lack of preparation and its delays in producing materials in the course of the verification proceeding also evidenced a failure to cooperate. In its verification report, Commerce noted that it requested cost and management reports, including the 1996 Budgeting Report, at the beginning of the verification. Heveafíl, however, waited until the third day of a scheduled five-day verification to inform Commerce that some of the reports no longer existed and that a copy of the 1996 Budgeting Report could not be located. On the fourth day, Heveafil located a portion of the Budgeting Report, but even then, as noted, Commerce in many instances was unable to reconcile costs for those items included in the portion of the Budgeting Report that was produced.
Commerce also experienced delays and difficulties resulting from the Heveafíl employees’ lack of familiarity with the company’s accounting records. Commerce noted that the staff assigned to the verification could not identify the appropriate officials to address particular topics and that photocopies of documents that Commerce had previously requested were not ready. Counsel for Heveafíl acknowledged at oral argument that Heveafíl had been subject to verifications in the first and second administrative reviews under the anti-dumping order, but argued that different employees had been involved in those reviews. Even so, Heveafíl should have known based on the company’s prior experience and the questionnaires it had received from Commerce that it was critically important for verification to retain source documents and have them ready and available for inspection. The failure to maintain source documents, including the 1996 Budgeting Report, and the lack of preparation and responsiveness constitute substantial evidence of Heveafil’s failure to cooperate to the best of its ability.
Heveafíl argues that in light of its substantial efforts to cooperate with the review, Commerce should have assigned it a lower dumping margin. Because of its cooperation, Heveafíl contends that Commerce should be limited to the highest margin Commerce has assigned to Heveafil, rather than the highest margin Commerce has assigned to any producer subject to the order. We disagree. While the antidumping statute distinguishes between respondents who have not cooperated and those who have, neither the statute nor the pertinent regulations address the weight *850to be given to different degrees of cooperation. Section 1677e(b) lists various sources on which Commerce may rely when using facts otherwise available, but it does not specify which sources are acceptable depending on the extent of an investigated company’s cooperation. While Commerce has on occasion assigned a lower antidumping duty margin to respondents who have cooperated to a significant degree, there is no established formula requiring less adverse margins when respondents have been partially cooperative. We have previously noted that “[i]n the ease of uncooperative respondents, the discretion granted by the statute appears to be particularly great, allowing Commerce to select among an enumeration of secondary sources as a basis for its adverse factual inferences.” Ta Chen Stainless Steel Pipe, Inc. v. United States, 298 F.3d 1330, 1338-39 (Fed.Cir.2002). In light of the broad discretion granted to Commerce to select among various sources for the appropriate adverse facts antidumping duty margin, we decline Heveafil’s invitation to hold that in light of Heveafil’s efforts to cooperate in the verification investigation, Commerce was barred from selecting a dumping margin greater than the largest margin imposed on Heveafil in any previous review.
C
The final question is whether Commerce may continue to use an invalidated antidumping duty margin as corroboration for an adverse facts available rate in its determination of a rate to be charged against Heveafil in the 1995-96 administrative review. In light of our decision in D & L Supply Co. v. United States, 113 F.3d 1220 (Fed.Cir.1997), we hold that it may not. Both parties agree with this conclusion.
In D & L Supply Co., we held it unlawful for Commerce to use an invalidated rate as the best information available (“BIA”) rate, which was the statutory precursor to the current adverse facts otherwise available rate. We held that “when the dumping margin on which the BIA rate is based is invalidated before the BIA rate has become final, it is irrational to ignore the invalidity of the underlying rate and to uphold the BIA rate as purportedly based on the ‘best information available.’ ” D & L Supply Co., 113 F.3d at 1224. The principles announced in that case apply similarly here.
Section 1677e(c) of the antidumping statute provides that “[w]hen the administering authority or the Commission relies on secondary information rather than on information obtained in the course of an investigation or review, the administering authority or the Commission, as the case may be, shall, to the extent practicable, corroborate that information from independent sources that are reasonably at their disposal.” Heveafil’s assigned rate was based on an antidumping duty margin assigned to Rubfil Sdn. Bhd. in the third administrative review under the order. Rubfil’s rate, however, was later invalidated in litigation and subsequently reduced to 36.14 percent. We have stated that Congress “intended for an adverse facts available rate to be a reasonably accurate estimate of the respondent’s actual rate, albeit with some built-in increase intended as a deterrent to non-compliance.” De Cecco, 216 F.3d at 1032; Ta Chen, 298 F.3d at 1340. In light of that statement and the principle set forth in D & L Supply, the invalidated Rubfil rate cannot serve as an acceptable basis for corroboration in the facts otherwise available analysis. The rate assigned to Heveafil was thus uncorroborated. Accordingly, we reverse Commerce’s reliance on Rubfil’s invalidated 54.31 percent dumping rate as *851the rate to be applied to products sold by Heveafil during the relevant time period. Commerce should proceed to assign a new dumping rate that is statutorily authorized and properly corroborated.
Each party shall bear its own costs for this appeal.