Floral Trade Council (FTC) (plaintiff) and Asociación Colombiana De Exportadores de Flores (Asocolflores) (defendant) both appeal from the judgment of the United States Court of International Trade entered July 22, 1993 (No. 90-06-00290). We affirm.
BACKGROUND
This court action commenced with FTC’s and Asoeolflores’s challenge to the final results of an International Trade Administration (ITA) antidumping duty review respecting certain fresh cut flowers from Colombia. 55 Fed.Reg. 20,491 (1990). FTC contested, inter alia, ITA’s use of constructed value for the foreign market value as opposed to using third country prices. Asocolflores objected, inter alia, to ITA’s selection of monthly instead of annual average U.S. prices.
The trial court upheld ITA’s use of constructed value, finding reasonable ITA’s interpretation of its governing statute that third country prices may be abandoned if there is an adequate factual basis in the record for doing so. Further, that court credited the factors underlying the ITA conclusion that using constructed value was proper because reliance on European prices for third country prices would produce misleading results due to differences between the floral markets in Europe and the United States. The trial court was also persuaded that monthly averaging of U.S. prices was an appropriate compromise, whereas annual averaging would mask dumping. Floral Trade Council v. United States, 775 F.Supp. 1492, 1497, 1500 (Ct.Int’l Trade 1991). After remand on grounds not pertinent to this appeal, judgment was entered upholding ITA’s review and the parties appealed to this Court. See Cabot Corp. v. United States, 788 F.2d 1539, 1543 (Fed.Cir.1986) (remand decision appealable only if rule of finality is satisfied).
ANALYSIS
The ITA determinations under review must be upheld unless the determinations are “unsupported by substantial evidence on the record, or otherwise not in accordance *1202with law.” Tariff Act of 1930, § 516A, 19 U.S.C. § 1516a(b)(1)(B) (1988);1 Rhone Poulenc, Inc. v. United States, 899 F.2d 1185, 1189, 8 Fed.Cir. (T) 61, 65 (Fed.Cir.1990). Substantial evidence “means such relevant evidence as a reasonable mind might accept as adequate to support a conclusion.” Mat-sushita Elec. Indus, v. United States, 750 F.2d 927, 933 (Fed.Cir.1984) (citing Universal Camera Corp. v. NLRB, 340 U.S. 474, 477, 71 S.Ct. 456, 459, 95 L.Ed. 456 (1951)). Moreover, under traditional principles of judicial deference to agency interpretation of statutes the agency administers, our review questions whether the agency’s interpretation is reasonable. Chevron U.S.A. v. Natural Resources Defense Council, 467 U.S. 837, 844, 104 S.Ct. 2778, 2782-83, 81 L.Ed.2d 694 (1984); Daewoo Elec. Co. v. International Union, 6 F.3d 1511, 1516 (Fed.Cir.1993).
I.
Constructed Value
On appeal, FTC argues that ITA unlawfully abandoned third-country prices as the basis for foreign market value. Citing statutes and regulations expressing a preference for third-country prices, FTC contends that ITA could not depart from that preference in this case. FTC asserts that verified third country prices were available and that no unusual facts justified ITA’s rejection of them.
The Tariff Act of 1930, as amended in 1979, directs ITA to determine the foreign market value based on (1) the price at which similar merchandise is sold in the country from which exported (here Colombia); or (2) if the exporter’s home market consumption is low or nonexistent, then the price at which the merchandise is sold for exportation to a third country. 19 U.S.C. § 1677b(a)(1). ITA’s own regulations announce a clear preference for third country prices if adequate and verifiable information is available. 19 C.F.R. § 353.48(b). However, the statute also gives ITA the discretion to construct a value when home market prices cannot be determined. 19 U.S.C. § 1677b(a)(2).
In the instant case, ITA determined that it could not determine the price under the first statutory directive and, due to extraordinary circumstances, it chose to use a constructed value rather than third country prices. ITA reasoned that the United States and European markets are not positively correlated and that this negative correlation could mask dumping in certain instances or exaggerate it in others. First, ITA determined that the European market for flowers is mature with a constant demand, whereas the United States market exhibits extreme volatility due to sporadic gift-giving. Second, Colombian growers are not a constant participant in the European flower auction houses and, therefore, Colombian growers sell flowers in Europe only occasionally. Third, European and United States holidays often do not coincide, resulting in different peak periods in the two markets. Last, Colombians plan their production cycles on the U.S. market and, therefore, sell excess flowers in markets where they did not necessary plan to sell, which adds an element of chance to prices obtained for those flowers.
FTC takes issue with ITA’s exercise of discretion in choosing to use a constructed value rather than using third country prices. We are unpersuaded of factual or legal error in ITA’s selection of constructed value or of an abuse of discretion. We find no restriction either in the statute or the regulations limiting use of constructed value to situations where verified information is not available for third country prices. Further, ITA’s explanation for rejecting such prices was reasonable.
FTC also attacks the evidence underlying the factors on which ITA relied. Both Aso-colflores and the government amply demonstrate that substantial evidence exists to support ITA’s findings. More than ten years ago, we faced a similar evidentiary challenge and stated:
That [FTC] can point to evidence which detracts from the evidence which supports the [ITA’s] decision and can hypothesize a reasonable basis for a contrary determination is neither surprising nor persuasive. It is not the function of a court to decide that, were it the [ITA], it would have made *1203the same decision on the basis of the evidence. Our role is limited to deciding whether the [ITA’s] decision is “unsupported by substantial evidence on the record, or otherwise not in accordance with law.”
Matsushita Elec., 750 F.2d at 936. “[T]he possibility of drawing two inconsistent conclusions from the evidence does not prevent an administrative agency’s finding from being supported by substantial evidence.” Id. at 933 (quoting Consolo v. Federal Maritime Gomm’n, 383 U.S. 607, 619-20, 86 S.Ct. 1018, 1026-27, 16 L.Ed.2d 131 (1966)). Here, FTC asks us to accept its view of the facts, which we decline to do.
Having considered all of FTC’s arguments, we affirm ITA’s use of constructed value on the facts of this case.
II.
Monthly Averaging
Asocolflores’s appeal challenges ITA’s selection of monthly, instead of annual, average U.S. prices as unsupported by substantial evidence. Asserting that ITA’s reasoning for selecting a monthly average is merely speculative, Asocolflores states that ITA’s choice is not based on any evidence and is inadequate as a matter of law. Asocolflores notes that ITA’s internal analyses all support an annual averaging. Further, Asocolflores claims that the constructed value for foreign market value is based on an annual average, so that comparing monthly U.S. prices to a constructed annual value will necessarily establish dumping anytime the Colombian grower does not meet its average annual profit. With perishable flowers, growers must occasionally sell at a loss, but at other times the growers sell at a great profit, so that their annual profit balances the peaks and valleys of selling. Last, Asocolflores notes that ITA’s reasoning for rejecting an annual average for U.S. prices directly contradicts its reasoning for rejecting third country prices for foreign market value.
The Tariff Act empowers ITA to use averaging in determining United States price. 19 U.S.C. § 1677f-1(b). The determination to select averages rests exclusively with the ITA if the averages are “representative of the transactions under investigation.” Id. Because ITA has discretion to use averages, Asoeolflores’s challenge must be directed to whether the averages are “representative” of the transactions under investigation.
In its preliminary determination, ITA found averaging was necessary to account for perishability. Growers do not control when flowers will be sold nor the price at which they will be sold. With flowers having such a limited life span, growers cannot await a better market by warehousing nor may growers regulate production in the short term. Growers must sell what their plants produce when the plants produce. In its final determination rejecting economic reports submitted by the growers, ITA stated that annual averaging would mask dumping by balancing high prices in peak months with low prices in other months. ITA also rejected FTC’s proposal to use a daily or weekly averaging because ITA wanted a long enough period to account for both distress and nondistress sale prices. Reasoning that flowers can last more than one week and that perishability is a function of variables other than merely time, ITA selected a monthly period for averaging.
The trial court noted: “No period accounts for the legitimate concerns of all parties. Monthly averaging is one acceptable compromise. ITA averaged over what it considered the shortest period possible- to ensure that dumping was not obscured entirely, and at the same time, to account for as many perishability factors as possible.” Floral Trade Council, 775 F.Supp. at 1500. The government argues in its brief that 19 U.S.C. § 1675(a)(2)(A) directs ITA to compare each entry of merchandise subject to the anti-dumping order, whereas Asoeolflores’s annual averaging would violate that mandate by masking individual instances of dumping. Under an annual averaging, dumping would only be found if the annual aggregate sales of a foreign seller were at less than fair value.
Based upon the particular factual record before us, we agree with the government that an annual average would mask dumping related to individual sales. Although some dumping would also be masked under a monthly average, Commerce chose to use a *1204monthly average as representative of the U.S. price to account for perishability of the flowers. Even Asocolflores admits that “in general, shorter averaging periods are preferable to longer averaging periods.” Basing the U.S. price on an annual average in this market would completely eviscerate determining dumping on the statutorily mandated “each entry of merchandise.” Yearly averaging is clearly not more “representative” of the U.S. price than monthly averaging in the circumstances of this case.
In sum, Asocolflores has not persuaded us that ITA abused its discretion, committed an error of law nor made its determination of the amount of dumping unsupported by substantial evidence.
AFFIRMED.
. All citations to the U.S.Code are to the 1988 version unless noted otherwise.