Utilizing an Almost Ideal Demand System

Transcription

1 792 Bulgaran Journal of Agrcultural Scence, 19 (No 4) 2013, Agrcultural Academy Analyzng UAE s Imports of Fresh Fruts Utlzng an Almost Ideal Demand System M. S. Gheblaw 1, M. K. Alashry 2, S. Sherf 1, A. Basarr 1 and Z. Ul- Haq 3 1 Unted Arab Emrates Unversty, Department of Agrbusness and Consumer Scences, UAE 2 Suez Canal Unversty, Department of Agrcultural Economcs, Egypt 3 Abdul Wal Khan Unversty, Department of Management Scences, Pakstan Abstract Gheblaw, M. S., M. K. Alashry, S. Sherf, A. Basarr and Z. Ul- Haq, Analyzng uae s mports of fresh fruts utlzng an almost deal demand system. Bulg. J. Agrc. Sc., 19: Ths paper provdes estmates of expendture and prce elastcty for mported fresh fruts n UAE usng a Lnear Approxmate Almost Ideal Demand System (LA/AIDS) over the perod Results showed that uncompensated or Marshallan own prce elastcty of mport demand have negatve sgns and nelastc except for grapes and oranges. All cross-prce elastcty among the sx commodtes s nelastc. Grape mports are substtutes for banana, date, mango and orange mports. Apple mports are substtutes for date and orange mports. Results also show that apple mports are complements to banana, mango and grape mports. Banana mports complement the mports of all the other fve fruts. However, there s no cross-prce elastcty between orange mports and date mports. All Compensated or Hcksan own prce elastcty have the expected sgns and nelastc except for grapes. All compensated own prce elastcty s relatvely nelastc except for grapes, whch s greater than one at Banana and dates are complementary goods wth mango and orange mports. Grape mports are substtute wth apple, banana, date, and mango mports. Apple mports are substtute for banana, date, grape and orange mports. From the expendture elastcty, both banana and grape mports appear to be luxury fresh fruts wth expendture elastcty greater than one. Other three fresh fruts are to be consdered necesstes as ther expendture elastcty are less than one whle date mports had negatve expendture elastcty. Key words: LA/AIDS model, mport demand, fresh fruts, uncompensated and compensated elastcty Introducton Unted Arab Emrates (UAE) s an mportant net mporter of fresh fruts among the members of the Gulf Cooperatve Councl (GCC) countres. The relatve mportance of UAE mports of fresh fruts for the years 1990, 2000 and 2007 are presented n Table 1. The total values recorded ndcate a steady rse for sx fresh frut mports (apples, bananas, dates, grapes, mangoes and oranges) from $106.6 mllon n year 1990 to $ mllon n year 2000 and to $ mllon n year These fgures represent 74.4 percent of total mports of all fresh fruts n 1990, 86.6 percent n 2000 and percent n year 2007, respectvely. In 1990, the expendtures of UAE on oranges, dates and apples mports were $24, $ 21.6 and $ 21 mllon comprsng 16.9 percent, percent and percent of total expendtures on all sx fresh fruts, respectvely. Meanwhle, the expendture values on bananas, mangoes and grapes mports occuped the fourth, ffth and sxth ranks wth ratos 9.8 percent, 9.1 percent and 8.45 percent of total expendtures on all sx fresh fruts respectvely, for the same year. An mportant pont to note s that the expendtures on apples, oranges, mangoes and grapes mports ncreased from $21, $ 24, $13 and $12 n year 1990 to $94.57, $46.47, $34.45 and $30.56 mllon n year 2007, respectvely. By contrast, the expendtures on bananas and dates mports declned from $14 mllon and $21.6 mllon n year 1990 to $3.62 mllon and $14.65 mllon n year 2007, respectvely. Fgure 1 represents the summary statstcs of expendture shares for the sx fresh fruts mports of UAE durng The expendture share of apple mports showed a gradual ncrease through the perod to 2007 from n 1990 to n E-mal: gheblaw@uaeu.ac.ae, prof_khary@hotmal.com, shern.sherf@uaeu.ac.ae, abasarr@uaeu.ac.ae, and zhaq@aup.edu.pk

2 Analyzng UAE s Imports of Fresh Fruts Utlzng an Almost Ideal Demand System 793 The expendture share of bananas llustrated steady ncrease wth n 1990 to reach a maxmum wth rato n 2001, and then declned to n The expendture share of dates reached ts maxmum n year 2005 wth and t was at a mnmum n year 2002 wth a share of Meanwhle these shares declned from n 1996 untl year 2004 wth a rato of The hghest expendture share for mango mports was n year 2005 wth a share of 0.223, t was at a mnmum of n year 1993.The expendture shares declned through years 1994 to 2001, and they ncreased from years 2002 to It can be observed from the same fgure that the expendture shares of oranges mports ranged between as upper lmt n year 2003 and as a lower lmt n year Moreover, these expendture shares declned for years 1991 to 1994, 1996 to 1999 and 2004 to As can be seen from Fgure 1, the expendture shares of grape mports reached a maxmum wth a share of n year 2000 and t was at mnmum n year 1997 wth a (Table 1 and Fgure 1). The remander of ths paper s organzed as follows. Secton 2 llustrates the data sources. Secton 3 s concerned wth theoretcal framework. In secton 4, the emprcal re- Table 1 Expendtures on fresh fruts mports of UAE for years 1990, 2000, and Import values, $ mllons % Import values, $ mllons % Import values, $ mllons (1) - Sx fresh fruts A. Apples B. Bananas C. Dates D. Grapes E. Mango F. Orange Total values for the mentoned sx fruts (2) - Other fresh fruts* Total Import values (1+2) *Other fresh fruts nclude pears, aprcots, avocados, Kk peaches and nectarnes, pneapples, grapefruts, ctrus fruts, tropcal fresh fruts,.etc Sources: Collected and computed from FAO database. % Apples Bananas Dates Grapes Mangoes Oranges Fg. 1. Expendture shares of UAE fruts mports ( )

3 794 M. S. Gheblaw, M. K. Alashry, S. Sherf, A. Basarr and Z. Ul- Haq sults are reported and dscussed. Secton 5 wraps up wth conclusons. Data Data used n ths study are readly avalable onlne. Annual tme seres data from 1990 to 2007 were obtaned from FAOSTAT. Data of mport values n mllon dollars were used as expendtures from UAE on sx fresh fruts: apples, bananas, dates, grapes, mangoes and oranges. The mport prces per ton n dollars for each commodty were calculated as mport values n mllon dollars dvded by the mport quanttes n tones. Theoretcal Framework The Almost Ideal Demand System (AIDS) was ntroduced by Deaton and Muellbauer (1980) and later ts varant the lnear approxmaton of AIDS (LA /AIDS). AIDS provdes a useful framework for modelng demand systems. It s based on economc theory and satsfes the axoms of choce. It can also be used to test homogenety and symmetry restrctons. The coeffcents n a LA/AIDS model are easy to estmate and nterpret. The AIDS model was appled n several studes to analyze the world mport demand for agrcultural products (Yang and Koo, 1994; Andayan and Tlley, 1997; Soshnn et al., 1999; Dameus et al., 2000; Schmtz and Seale, 2002; Poudel and Kethy, 2008; and Nzaku and Houston, 2009). In order to provde emprcal measures of the senstvty of the demand for mported fresh fruts n UAE to changes n prces and ncome, the paper concentrated on the estmaton of demand equatons system by applyng a dynamc LA-AIDS to data on mports of sx fresh frut types over the perod of The general specfcaton of the AIDS model ndcates that the demand for mported fresh fruts n UAE can be gven by:, (1) where S t represents the expendture share of the mport good n the perod t; P jt represents the mport prce of the j th good n tme t; X represents the total expendtures on all fresh frut mports ncluded n the system n tme perod t; ln P s translog prce ndex; α jt, β t and γ j are parameters assocated wth the system and µ s the error term. Deaton and Muellbaur used a translog prce ndex (P*), whch makes the demand system non-lnear as follows, (2) Ths prce ndex makes the demand non-lnear, whch normally complcates the estmaton process. To overcome the non-lnearty problem, Deaton and Muellbaur modfed the translog prce to be approxmated usng a Stone prce ndex to replace the translog prce ndex that t can be expressed as: ln P = Σn 1 S t ln P, (3) j Ths substtuton wll cause a smultanety bas (measurement error), because the dependent varable (S t ) also appears on the rght hand sde. Hence followng Eales and Unnevehr (1988) and to avod ths problem, we use the lagged expendture share (S t ) n Stone s geometrc prce ndex by one perod (S t-1 ) for equaton 3, and then replacement of equaton (3) wth the lagged dependent share nto equaton (1) yelds the LA/AIDS, defned as follows: S t = α 1 + Σ n j=1 γ j ln P j + β t (ln(x) - Σ n =1 S ln P ) + u, (4) t-1 j Equaton (4) of LA/AIDS can be used n estmatng the parameters of the mport demand system equatons for sx fresh fruts n UAE and then they can be used to compute the own prce, cross- prce and expendture elastcty. The regularty condtons mpled by expendture constrants and utlty maxmzaton mpose the followng restrctons to the system: n n n Addng up: Homogenety: Symmetry: γ j = γ j ; j for any two commodtes. The elastcty has been computed at the sample means of expendture shares. The uncompensated (Marshallan) ownprce elastcty (ε ) and cross-prce elastcty (ε j ) are derved from Alston et al. (1994) as follows: ε = S, (5) S j j, j, (6) S S The formula used to calculate the expendture elastcty could be wrtten as: S 1 n 1 j 0 1, (7) S A postve value shows that commodty s normal. The ncome compensated (Hcksan) own-prce elastcty ( ) and cross-prce elastcty ( j ) respectvely are obtaned by applyng the Slutsky decomposton to (7) and usng the prce ndex n equaton (3). These can be wrtten as: 1 1 j

4 Analyzng UAE s Imports of Fresh Fruts Utlzng an Almost Ideal Demand System 795 j j S j S, j (8) The consumer theory suggests that compensated ownprce elastctes are negatve for normal goods. Moreover, f (6) and (8) are postve, the two goods are substtuton, otherwse they are complements. Emprcal Results Ths secton s dvded nto two subsectons; frst s concerned wth the test results of the demand system model. Second s focused on parameter and elastcty estmates of the model. All the estmates of the model usng Zellner s (1962) teratve restrcted seemngly unrelated regresson (IRSUR) for total of 18 observatons, the perod to 2007 wth E- vews 7 software package. Results of the tests Ths subsecton shows the results for the followng tests: (1) the structural break, (2) unt root, and (3) co-ntegraton n the resduals of the demand system model. Structural break The frst test s concerned wth the structural breaks n the mport demand equatons for sx fresh fruts. The structural breaks have taken place when a change s notced n the model parameters. A seres, whch was specfed as non-statonary n the absence of structural break, was actually trend statonary once structural break was computed n the regresson parameters of the model (Jha and Sharma, 2001). In order to fnd out years n whch structural breaks appear a set of resduals from the ftted LA/AIDS expendture shares (equaton 4) were examned. A structural break occurs durng the perod(s) where the resduals exceeded two standard devatons (Taljaard et al., 2003). Table 2 presents the results of structural breaks of the expendture shares for sx fresh frut mports by UAE for years The maxmum structural breaks occurred for dates n 1993 by and for oranges n 2004 by It s nterestng to observe from the same table that all values are less than two. Ths means that there are no ndcatons of structural breaks from Unt root test A tme seres s sad to be a statonary process f ts mean and varance are both constant over tme. Ths means that the seres s not a random walk and t has no unt root. By contrast, a tme seres s descrbed as beng a non-statonary, whch t s a random walk, and t has a unt root. Accordngly, statonary of a tme seres s mportant because t can nfluence ts behavor. If a seres s statonary wthout any dfferencng, t s ntegrated order zero, d(0). On other hand, a tme seres that has statonary frst dfference s desgnated of order one, d(1). Therefore, any tme a seres s used n econometrc analyss, t must be statonary (Granger and Newbold, 1986). Table 2 Structural breaks of the expendture shares for sx fresh frut mports by UAE for years Years Apples Banana Dates Grapes Mangos Oranges

5 796 M. S. Gheblaw, M. K. Alashry, S. Sherf, A. Basarr and Z. Ul- Haq It s well known that most economc varables are non-statonary and the presence of unt root n the resduals may be nvald asymptotc dstrbutor of the estmators. Therefore, the expendture shares dealt wth the demand system model need to be tested for unt root. In ths paper, Augmented Dckey Fuller (1979 &1981) s used to test for the unt root resduals for demand model system (equaton 4) as follows: y t k t 1 + d j yt j + eε t j= 1 = c + Τ + αa y for j = 1,2 k (9) Ths augmented specfcaton s used to test the followng hypothess: The null hypothess (H 0 ) α= 0; and he alternatve hypothess (H a ) α<0. If we could not reject the null hypothess, ths means that seres α contans a unt root; where α= p-1 and calculated usng conventonal t-rato for α. The order for Augment Dckey Fuller (ADF) tests were ascertaned on mnmum Akake Informaton Crteron (AIC). Table 3 presents the ADF test results for both the level seres and frst dfference of the varables used to estmate the expendture share equatons for fresh frut mports by UAE when a tme trend and no tme trend are ncluded n the mport demand system model. As may be seen n Table 3, an mportant result obtaned by ADF s to reject the null hypothess at the level seres and all the varables used to estmate the demand system are generated by a unt root process except mport prce of banana when a tme and no tme trend are ncluded n the model. Ths means that the tme seres s a non-statonary when tested at level seres. Once we fal to reject the null hypothess, we wll carry out the ADF test of the seres statonary at ts frst dfferenced value. It can be observed that the tme seres data of all the varables used to estmate the mport demand equatons at the frst dfference were the stronger the rejecton of null hypothess of unt root when a tme trend and no tme trend are ncluded n the model. Ths means that no unt root s present. In other words, all the varables used to estmate the demand system are statonary. Co-ntegraton test Testng of co-ntegraton s necessary to examne whether sx expendture shares are jontly determned wth ther respectve prces. Ths s because the lterature ponts out the possblty of a lnear combnaton of non-statonary varables mght be statonary. Once the expendture shares are ntegrated of the same order, d(1) wth the exogenous varables, then co-ntegraton can be establshed. In ths paper, two resduals-based tests are used to test for co-ntegraton namely; Augment Dckey Fuller (ADF) and Phllps-Perron (PP) tests. Phlps-Perron (1988) estmated the followng equaton, (10) Table 3 ADF test results for Expendture shares, mport prces and total expendtures Seres Level seres Frst dfferences No trend trend lags No trend trend Expendture shares Apple Banana Dates Grapes Mango Oranges Logarthm of prces Apple Banana Dates Grapes Mango Oranges Total expendture Crtcal values The lag length n the ADF test regresson s determned by Akake Informaton Crteron (AIC). I(d)

6 Analyzng UAE s Imports of Fresh Fruts Utlzng an Almost Ideal Demand System 797 where T s number of observaton and N s number of varables. The null hypothess s β 1 = 0versus the alternatve hypothess of β 1 < 0. Two versons of ADF and PP regressons are estmated wth ntercept only (β 2 = 0) and tme trend and ntercept.ths study uses ITSUR to get the resduals and test for statonary of the entre varables n the mport demand system model. Ths s due to fact that usng OLS does not control for correlaton of errors across the equatons and the results could lead to wrong mplcatons for ADF and PP tests. Table 4 ndcates the results of the two alternatves tests: ADF and PP for co-ntegraton between the expendture shares and the correspondng exogenous varables. The results of both ADF and PP tests reject the null hypothess of no co-ntegraton for sx expendture shares at fve percent sgnfcant level when a tme and no tme trend are ncluded n the model. Hence, these results for co-ntegraton suggest that the expendture shares are co-ntegrated wth ther exogenous varables. Ths means that all varables n the demand system are statonary. We can conclude from the prevous results that no ndcaton of structural break s present. At the same tme, unt root and co-ntegraton tests reveal that the seres are statonary and co-ntegrated wth ther explanatory varables. These results confrm that we do not need to use an error correcton mechansm (ECM) n ths study. Therefore, the mport demand system for sx fresh fruts by UAE (Equaton 4) s consdered vald to estmate the parameters. Parameter and Elastcty Estmates Ths subsecton contans two parts: frst ncludes parameter estmates and the second s concerned wth the elastcty estmates. Parameter Estmates The mposton of homogenety and symmetry condtons on the system may generate postve seral correlaton Table 4 Co-ntegraton test for mport demand seres Dckey-fuller Seres Co-ntegraton test n the resduals. Furthermore, the varance covarance matrx of resduals play an mportant role n the estmaton snce symmetry nvolves cross- equaton restrctons. To avod a sngular varance covarance, one of sx share equatons (Grapes n ths case) can be arbtrarly deleted and a system estmator can be appled to the remnder equatons of the system (only fve equatons) by usng IRSUR procedure and the parameters of the dropped equaton may be estmated by usng the addng-up property. The estmaton results for the lnear approxmate form of the Almost Ideal Demand System (LA/AIDS) for the mport demand for sx fresh fruts n the Unted Arab Emrates market are presented n Table 5. Values of R 2 ranged from 69 percent for apple as an upper lmt and 28 percent for mangoes as a lower lmt. About 60 percent of the estmated coeffcents are statcally sgnfcant at the one percent, fve percent and ten percent levels. The expendture coeffcents for UAE mports of fresh fruts ndcate whether commodtes are necesstes or luxures. However, all statstcally sgnfcant expendture coeffcents for UAE mports of fresh fruts wth the excepton of bananas are negatve. In addton, all expendture coeffcents are sgnfcant at the one percent, fve percent and ten percent levels except for mangoes. Furthermore, the coeffcents of grapes equaton were calculated from the addng-up condton wthout any sgnfcance. Elastcty Estmates Researchers are mostly nterested n the demand elastctes. The flexble functonal form of the LA/AIDS model allows us to easly carry out the elastcty analyss. The demand elastctes are calculated as functons of the estmated parameters and they have standard mplcatons. Accordng to Green and Alston (1990), elastctes n LA/AIDS model can be expressed as expendture elastcty as well as uncompensated and compensated own prce and cross- prce elastctes. The uncompensated and the compensated elastctes of Phlps-Perron Co-ntegraton test No Trend Trended No Trend Trended Expendture shares Apple Banana Dates Grapes Mango Oranges % crtcal

7 798 M. S. Gheblaw, M. K. Alashry, S. Sherf, A. Basarr and Z. Ul- Haq demand for sx fresh fruts mports by UAE were computed at sample means. Uncompensated elastctes By usng the estmated coeffcents n Table 5 and equatons 5 and 6, the uncompensated or Marshallan own and cross-prce elastctes are lsted n Table 6. Uncompensated own prce elastctes for sx fresh frut mports by UAE have negatve sgns as expected accordng to economc theory. The own prce elastcty for fresh grapes mport s -1.6 and slghtly more than untary for fresh oranges,.e , mplyng that ther elastctes are elastc. Ths means that both fresh grapes and oranges mports are very senstve to mport prces changes. These results are also mportant for exporters snce a one percent ncrease n own prce would decrease mport demand both grapes and oranges by 1.6 and 1.02 percent, respectvely. Accordngly, a prce ncrease for these fruts ceters parbus would decrease total revenue for traders. By contrast, own prce elastctes for apples, bananas, dates and mangoes mports are nelastc, whch mples a one percent ncrease n own prce would decrease mport demand by 0.53, 0.82, 0.92 and 0.39 percent, respectvely. Thus, a small ncrease n own prce would ncrease total revenues from apples, bananas, dates and mangoes sales. Turnng to the cross-prce effects as shown n the same table, all crossprce elastctes among the sx commodtes are nelastc. It s also mportant for frut exporters to know the effects on ther products demand from prce changes of other competng fruts. Postve cross-prce elastctes show that two commodtes are substtutes, whereas negatve elastctes ndcate complementary products (Schmtz and Seale, 2002). The results show that grape mports are substtutes for banana, date, mango and orange mports. Apple mports are substtutes for date and orange mports. Results also show that apple mports are complements to banana, mango and grape mports. Banana mports complement the mports of all the other fve fruts. However, there s no cross-prce elastcty between orange mports and date mports. Compensated elastctes Compensated or Hcksan elastctes are calculated by usng the parameter estmates n Table 6 and formula 8 whle Table 5 Parameter estmates of LA/ AIDS model for years to Apples Banana Dates Mangos Oranges Grapes Apples (6.60***) Banana (-2.50**) (2.55**) Dates Mangos (3.75***) (-1.98*) (-1.38) Oranges (2.03*) (-2.33**) (2.29**) (-0.57) Grapes (-1.584) (-0.35) (2.29**) Expendture (-2.91**) (5.83***) (2.17*) (-1.2) (-2.66**) Constant (3.68***) (-5.64***) (2.29**) (2.97**) R DW EL SL AIC Note: t-rato are n parentheses where *, ** and *** denote sgnfcant at 0.10, 0.05 and 0.01 respectvely. The t-rato does not appear for coeffcents whch have been obtaned usng relevant restrctons. EL refers to equaton log-lkelhood, SL s the system log-lkelhood, and AIC s system Akake nformaton crteron.

8 Analyzng UAE s Imports of Fresh Fruts Utlzng an Almost Ideal Demand System 799 the expendture elastctes are calculated utlzng formula n equaton 7. The compensated own and cross- prce and expendture elastctes are ndcated n Table 7. Consstent wth economc theory, as for the case of uncompensated, all estmated own prce elastctes of sx fresh frut mports by UAE have the expected negatve sgns as pror. All compensated own prce elastctes are relatvely nelastc except for grapes, whch s greater than one at Ths means that mport demand of grapes s very senstve to mport prce changes. It s clear from the same table that all cross-prce elastctes among the sx fresh frut mports are nelastc as n the case of uncompensated. The estmated cross-prce elastctes show mostly the expected relatonshps of mport demand for pars of sx fresh fruts. Based on the negatve sgns of crossprce elastctes both banana and dates are complementary goods wth mango and orange mports. Other complementary fresh frut nduce apple mports wth mangoes mports. Results also ndcate that some goods are substtutes. It s clear from the table that grape mports are substtute wth apple, banana, date, and mango mports. Apple mports are substtute for banana, date, grape and orange mports. The expendture elastctes The expendture elastctes of the mported demand for all the fresh fruts by UAE are postve wth excepton of date mports (Table 7). Both banana and grape mports appear to be luxury fresh fruts wth expendture elastctes greater than one. Other three fresh fruts are to be consdered necesstes as ther expendture elastctes are less than one. Negatve expendture elastcty of date mports s because there are annually sgnfcant ncreases n UAE dates producton from 141 thousand tons n 1990 to reach 290 thousand tons n 1998, and then approxmately doubled to 536 thousand tons n Snce 2000 to 2007, UAE dates producton ranged between 750 thousand tons as lower lmt n 2005 and 760 thousand tons n 2004 as upper lmt. At the same tme, self-suffcency of UAE dates snce 1996 to 2007 was more than 100 (FAO). Self-suffcency s equal to domestc producton dvded by domestc consumpton. Accordngly, domestc consumpton s equal to domestc producton plus mports mnus exports. Therefore, expendtures on date mports had a negatve sgn due to avalablty of the product domestcally. Conclusons Ths paper estmated a demand system for mported fresh fruts n UAE usng a Lnear Approxmate of Almost Ideal Demand System (LA/AIDS) over the perod Results showed that uncompensated and compensated own prce elastctes for sx fresh frut mports have negatve sgns and nelastc except for uncompensated elastctes for grape and orange mports and compensated elastcty of grape mports Table 6 Uncompensated (Marshallan) Own prce and Cross prce elastctes of LA/AIDS Model Apple Banana Dates Mango Oranges Grapes Apple Banana Dates Mangos Oranges Grapes Note: the bold values are the own-prce elastctes, otherwse are the cross-prce elastctes Table 7 Compensated (Hcksan) own and cross prce elastctes and Expendture elastctes of LA/AIDS Model Apples Banana Dates Mangos Oranges Grapes Apples Banana Dates Mangos Oranges Grapes Expendture Note: the bold values are the own-prce and expendture elastctes, otherwse are the cross-prceelastctes.

9 800 M. S. Gheblaw, M. K. Alashry, S. Sherf, A. Basarr and Z. Ul- Haq whch were greater than unty. These results are mportant for exporters of these fruts because they llustrate whether or not an own prce change would decrease or ncrease revenues. Moreover, f the prces of fresh frut mports ncrease by a certan percentage, grape and orange mports would declne by more than the percentage ncreases n ther prces. Accordngly, lowerng the prces of both grapes and oranges exports to UAE would ncrease total revenue from both. By contrast, apple, banana and mango mports would drop by less than the percentage ncreases n ther prces. The results also ndcated that uncompensated and compensated cross-prce elastctes among the sx commodtes are less than untary n absolute values. Banana mports seem to be complementng the mports of the remanng fruts. On the other hand, grape mports seem to be substtutes for mports of the other fruts except for apples. Last fndng of ths emprcal analyss was that both banana and grape mports are consdered luxury fresh fruts wth expendture elastctes greater than one whle apple, mangoe and orange mports are necesstes as ther expendture elastctes are less than one. References Alston, J. M., Foster K. A. and Green R. D., Estmatng elastctes wth the Lnear Approxmate Almost Ideal Demand System: some Monte Carlo results. The revew of Economcs and Statstcs, 76 (2): Andayan, S. R. M. and Tlley, D. S., Demand and compettor among supply sources: The Indonesan frut mport market. Journal of Agrcultural and Appled Economcs, 29: Dameus, A., Tlley, D. S. and Brorsen, B. W., Carbbean demand of U.S. and rest-of-the-world starchy food (wheat, rce, corn and fresh potatoes): A restrcted source-dfferentated almost deal demand system. A. E Stllwater, OK., Oklahoma State Unversty, Department of Agrcultural Economcs. Deaton, A. and Muellbauer, J., An Almost Ideal Demand System. The Amercan Economc Revew, 70 (3): Dckey, D. A. and Fuller, W. A., Dstrbuton of the estmators for autoregressve tme seres wth a unt root, Journal of Amercan Statstcal Assocaton, 74: Eales, J. S. and Unnevehr, L. J., Demand for beef and chcken products: Separablty and structural change. Amercan Journal of agrcultural Economcs, 70: FAOSTAT, Food and Agrculture Organzaton of the Unted Natons at: Granger, W. J. and Newbold, P.,1986. Forecastng Economc Tme Seres. Second edton, Orlando, Florda, AcademcPress. Green, R. and Alston, J. M., Elastctes n AIDS Models. Amercan Journal of Agrcultural Economcs, 72: Jha, R. and Sharma, A., Structural breaks and unt roots: A further test of the sustanablty of the Indan fscal defct. ASARC Workng Papers , Australa South Asa Research Center, Australan Natonal Unversty, Canberra. Nzaku, K. and Houston, J. E.,2009. Analyss of U.S. Demand for fresh Tropcal Frut and Vegetable Imports. Annual Meetng, January 31- February 3, 2009, Atlanta, Georga 46850, Southern Agrcultural Economcs Assocaton. Nzaku, K. and Houston, J. E., Dynamc estmaton of U.S. Demand for fresh vegetable mports. The agrcultural & Appled Economcs Assocaton, AAEA&ACCI Jont Annual Meetng, Mlwaukee, Wsconsn, July, 26-29, Nzuma, J. M. and Sarker, R., An error corrected Almost Ideal Demand System for major cereals n Kenya. Amercan Agrcultural Economcs, Assocaton, Annual Meetng, Orlando, Florda, July 27-29,2008. Poudel, P. and Kethy, Jr, W., Analyss of Unted States and European mport demand for Shrmp. The Southern Agrcultural Economcs, Assocaton Annual meetng, Dallas, TX, February 2-6. Phllps, P.C.B and Perron, P., Testng for a unt root n tme seres regresson. Bometrka, 75: Schmtz, T., and Seale, J. L., Import Demand for Dsaggregated Fresh Fruts n Japan. Agrcultural and Resource Economcs Revew, 34 (3): Soshnn, A., Tomek, W. and Gorter, H., Elastctes of Demand for Imported Meats n Russa. Workng Paper 19, Department of Appled Economcs and Management, Cornell Unversty, Ithaca, NY. Taljaard, R. R., Alemu, A. G. and Scalkwyk, V. H. D., A lnearzed Almost Ideal Demand system (AL/AIDS). Estmaton of demand for meat n South Afrca (AEASA), October 2-3, 2003, Pretora, South Afrca Yang, S. R. and Koo, W. W., Japanese meat mport demand estmaton wth the source-dfferentated AIDS model. Journal of Agrcultural and Resource Economcs, 19: Zellner, A., An effcent method of estmatng seemngly unrelated regressons and tests for aggregaton bas. Journal of the Amercan Statstcal Assocaton, 57 (298): Receved October, 28, 2012; accepted for prntng February, 2, 2013.