VERSLAS: TEORIJA IR PRAKTIKA / BUSINESS: THEORY AND PRACTICE

Logistics has been identified as an area to build cost and service advantages. Therefore, companies are more focused on customer needs and trying to find ways to reduce costs, improve quality and meet the growing expectations of their clients. Indeed, the global competition has led managers to begin to address the issue of providing more efficient logistics services. In this regard, this paper presents an empirical study on logistics performance and global competitiveness. This paper also identifies the associations between logistics performances and global competitiveness. The results indicate that some variables in global competitiveness scores contribute much higher to the logistics performances than the other variables through analysis. The variables that contribute much higher than the other variables to the logistics performances are shown.


Introduction
Turkey has experienced dramatic economic growth over the past decade, and its economic outlook remains posi tive despite the impact of the global economic crisis. A stable government and continuous structural reforms have encouraged a stronger banking sector and stronger fiscal discipline, and these measures have transformed the eco nomy of Turkey, which needed to receive assistance from the International Monetary Fund approximately 10 years ago, to an economic powerhouse regionally and globally. In 2011, Turkey's GDP grew by approximately 8.5%, one of the highest rates of growth in the world (see Figures 1, 2 and Table 1). Turkey's economy is slowed to 2.2% growth in 2012. In 2013, the economy is expected to expand by 3%.
Although Turkey has low public debt, strong fiscal dis cipline, and falling unemployment, the rapid growth of its economy and increased domestic demand have led the cur rent account deficit to soar to approximately 10% of GDP. Chronic dependency on energy imports is a major reason for this ballooning current account deficit, an imbalance that is seen as a sign of an overheating economy (Acaravci and Ozturk 2009).
Consequently, the current account deficit remains the main liability of the Turkish economy. For instance, even though the International Monetary Fund revised its forecast for Turkey's current account deficit to 8.8% of GDP for 2012 from a September estimate of 7.4%, it is expected to reach a ratio of more than 9%. Further, despite seeming to be well equipped to withstand any future economic storm, Turkey's current account deficit must be reduced in the coming pe riod through a longterm strategy to reduce its dependence on imported energy, which costs the country approximately $50 billion per annum, and imports of intermediate goods that feed its industrial production. Moreover, analysts agree that the current account imbalance is unsustainable in the long run, especially in a country such as Turkey that is fi nanced primarily by shortterm inflows.
In order to apply an exponential trend model to fore cast the future values of GDP per capita for Turkey, the present study uses an autoregressive integrated moving average (ARIMA) model for a dataset that covers 12 time periods (2000)(2001)(2002)(2003)(2004)(2005)(2006)(2007)(2008)(2009)(2010)(2011). The model assumes that a parametric model that relates the most recent data values to previous data values and previous noise provides the best forecast for future data (see Figure 2 and Table 2). The outlook of Turkey's economy is depicted using GDP forecasts based on past GDP values. By considering such a GDP indicator, forecasting techniques can thus be applied to predict future GDP values.
For periods beyond the end of the time series, Table 2 shows the upper limits for the forecasts at a 95% confidence interval, assuming the fitted model is appropriate for the data.
The Turkish public seems to be increasingly skeptical about the EU in general and the country's accession in par ticular. In November 2006, one study showed that after cri ticism from the EU regarding the continued lack of reforms in Turkey, only a third of the Turkish population supported Turkey joining the EU (Yildiz et al. 2008). As stated by Yildiz et al. (2008), Turks are also beginning to view the EU in a more critical light, accusing member states of hypocrisy, double standards, and discrimination. In addition, Lejour and de Mooij (2005) claimed that although the country's accession generates economic benefits for Turkey, it does not greatly influence EU member states or countries in Central and Eastern Europe.
The remainder of this paper is organized as follows. Section 2 reviews the literature on Turkey's accession to the EU and discusses the strategic importance of Turkey. Section 3 introduces the data and methods used for the cluster analysis and presents the empirical findings. The paper is concluded in Section 4.

Turkey and the EU
In 1959, Turkey applied for associate membership of the European Economic Community (now the EU), which was granted four years later. However, its application for full membership was not made until 1987 because of poli tical developments in Eastern Europe and the Soviet Union (Hoekman, Togan 2005). Nas (2008) stated that Turkey's application for full membership has been met with conti nued opposition from most European governments and the European public. Nevertheless, on October 3, 2005, EU foreign ministers reached an historic agreement on a framework of accession negotiations with Turkey, which marked the official opening of negotiations towards full membership (Nas 2008). According to Nas (2008), full accession talks are expected to last much longer than those held with other EU member states because of the geographical circumstances peculiar to Turkey. The country is located between the Black Sea to the north, the Aegean Sea, Greece, and Bulgaria to the west, the Mediterranean Sea, Cyprus, Syria, Iran, and Iraq to the south, and Georgia, Armenia, and Azerbaijan (Nahcivan) to the east. This unique location provides different political and strategic challenges (Kalaycioglu 2005). Rabasa (2008) emphasized that Turkey influences the interests of western (particularly US) security in an area where the Middle East, the Balkans, and the Caucasus intersect. Moreover, Rabasa (2008) acknowledged the additional importance of Turkey because of the coexistence of Islam with secular democracy, globalization, and modernity.
This view of the geopolitical and strategic importance of Turkey is widely shared by European leaders. At a meeting of EU heads of state and government members, Turkey's strategic importance was underlined by statements by the Council (Kramer 2000). Indeed, Turkey's clear eligibility for accession to the EU has defined how EU institutions, particularly the European Parliament, have approached the country in the past decade (Collins 2003). Tiersky (2010) acknowledged that viewing Turkey as a potential member state led to congruence between the strategic interests of the EU and its normative goals. In addition, Tiersky (2010) stated that Turkey has always been crucial to European se curity interests, both during the Cold War, when it stood as a bulwark against Soviet expansionism, and subsequently as a beacon of democracy (Tiersky 2010). In fact, Turkey has for centuries been a part, and sometimes a critical part, of the political, economic, and security system in Europe (Khalilzad 2000). With one of the few growing economies in the region and a renewed assertiveness locally (Whitman, Juncos 2011), Turkey's role as a regional power has also become more important in recent years. For example, it has played an important role as a mediator between Syria and Israel and in the Middle East peace process (Whitman, Juncos 2011). Further, as Turkey has become more capable and assertive in diplomatic, economic, and military terms, it has developed into a more important strategic partner for the west in the troubled regions of the world, from the Balkans to Central Asia and the Middle East (Khalilzad 2000). In summary, the size, location, and strategic impor tance of Turkey distinguish it from previous EU accession candidates (Tiersky 2010).
Politically, based on the size of the Turkish population, there have been some concerns that Turkey's influence on decisionmaking in European institutions would be overly significant (Nas 2008). In addition, other issues have also lead to protracted negotiations (Nas 2008). These issues include the fear of the increased migration of workers from Turkey, the alignment of key sectors in areas such as agri culture, transport, energy, and the environment, and the management of the EU's new external borders. According to Tiersky (2010), European policymakers are worried that the accession of Turkey would affect the market for goods and services, labor markets, the budgetary projections in the EU, and EU foreign policy. Aydin and Acar (2010) stated that the possible enlar gement of the EU to include Turkey is not just a political issue, but also has two important economic consequ ences for both the EU and the country. First, the large disparities in real wages between the two regions offer incentives for Turkish labor to migrate to the EU. Second, the regional mobility of capital would cause the rate of return values to converge towards growth rates at the steady state. Kibris and MüftülerBac (2011) expressed that, especially for the least developed countries on the continent, the process of EU enlargement, which has led the EU to double its size in the past six years, is testament to its increasing attractiveness. However, with each new enlargement, EU accession has become more difficult. In addition, openended accession is a new mechanism for integrating a candidate country (Ugur 2010). As argued by Ugur (2010), unlike previous enlargements, however, this mechanism does not specify the deadline for becoming a member and may suggest important exemptions that restrict the membership entitlements of new members. Consequently, current and future candidates may be less committed to regional reforms, while the EU may delay before deciding the date of accession and the entitlement package associated with it.
Moreover, despite the importance of Turkey in terms of foreign policy and its compliance with membership criteria, recent prospects for Turkish membership have worsened (Gerhards, Hans 2011). The progress of Turkey's accession to the EU slowed in 2010 because of its persis tent refusal to implement the Additional Protocol to the Association Agreement on the access of Cypriot ships and planes to Turkish ports and airports (Whitman, Juncos 2011). Turkey's accession also depends on the internal dy namics of the EU and the willingness of the EU to accept Turkey (MüftülerBac 2008). In addition, as acknow ledged by Gerhards and Hans (2011), most citizens of EU member states are opposed to Turkish membership, believing that Turkey would be a powerful player in the future of the EU because of its large population (Pahre, UcarayMangitli 2009). Others argue that this degree of power would negatively affect the EU. Pahre and Ucaray Mangitli (2009) explored whether these claims were inf luenced by the spatial models of EU policies and found that the preferences of Turkey are sufficiently outside the mainstream of the EU to have little influence on daily policy under agreement, codecision, consultation, and cooperation procedures. Alba and Park (2005) proposed that Turkish members hip would remain one of the most important issues facing the EU in coming years. The authors also highlighted the substantial opposition to Turkish entry because of the large income gap between Turkey and the EU and the economic costs that Turkish membership may impose on existing members. This implies that the narrowing of the income between Turkey and the EU over time would weaken the opposition to Turkish membership and vice versa (Alba, Park 2005). As noted by Lammers (2006), from a macroe conomic point of view, full membership for Turkey would be an advantage: where incumbent states are concerned, the disadvantages in terms of their GDP values seem to be low. In other words, Turkey's accession to the EU brings about economic benefits for Turkey without significantly affecting EU member states or countries in Central and Eastern Europe (Lejour, de Mooij 2005).

Trade, the EU, and Turkey
Indeed, it has been argued that the accession of Turkey would improve the welfare of EU members. For example, EU countries would benefit from welfare gains from com parative advantages and the growth effects of integration, while the migration of Turkish labor to the EU would affect overall EU welfare (Togan 2004). Some Turkish sectors such as textiles would also expand considerably at the expense of those sectors in Central and Eastern Europe (Lejour, de Mooij 2005). Further, the greatest economic gains would probably be achieved through reforms to national institu tions in order to improve the functioning of the public sector and ensure transparency for investors and traders (Lejour, de Mooij 2005).
Through the harmonization of trade legislation, EU companies would be able to use Turkey as a joint inves tment and as an export base for the Middle East and Eurasia (Togan 2004). Istanbul is fast becoming a location for the headquarters of transnational corporations for operations in the Caucasus and Central Asia. The EU would thus benefit from increased trade in the region (Togan 2004). Finally, Turkish membership would help ensure stability and security in the Balkans and Caucasus, allowing the EU to increase its energy security and reduce defense spending (Togan 2004).
Trade openness in Turkey has been rising in recent de cades, especially after its import substitution policies were abandoned in favor of measures of trade integration in the 1980s (Antonucci, Manzocchi 2006). The sum of exports and imports as a percentage of GDP increased from 18% in 1980 to 48% in 2001, when a severe financial crisis led to a contraction of 7.5% of GDP, after a period of sustained growth (Antonucci, Manzocchi 2006). Nevertheless, the EU remains by far the largest trading partner of Turkey, accoun ting for just over half of its exports and just under half of its imports. Since 1963, the EU has granted Turkey preferential trade status, with the Ankara Association agreements lea ding to a gradual reduction in import tariffs (especially on the EU side), the adoption of parts of the regulatory body of the EU (the socalled acquis communautaire) by Turkey, and the provision for the gradual creation of a customs union, which finally, after several delays, became operational in 1996. Lejour et al. (2004) and Flam (2003) have both analy zed the likely impact on trade by Turkey's accession into the EU. Lejour et al. (2004) estimated a gravity model of trade flows for a wide range of countries including goods and services trade and found that the EU model must be included in sectoral regressions to account for the particu lar intensity of trade relations between EU members (see also Antonucci, Manzocchi 2006). Although the empirical results differed considerably across sectors, the authors estimated that averageweighted sectoral EU-Turkey bilateral trade could increase by 34% if Turkey were an EU member. By adopting a different framework, Flam (2003) highlighted an even greater impact of the acces sion of Turkey's overall trade volume with the EU (over 46%). Moreover, Philippidis and Karaca (2009) found that common budget transfers would significantly influence the economic conditions of any hypothetical scenario of membership and concluded that EU membership would undoubtedly benefit Turkey. Through governmental stability, numerous constitutional and judicial reforms have been approved for Turkey's eventual membership (Philippidis, Karaca 2009).

Data
World Bank data were used as the basis of the presented cluster analysis. These annual data comprise a comprehen sive collection of countryspecific economic and develo pment indicators countries. From the database, selected important indicators were taken into consideration under the following major titles: the environment, economic po licy and debt, financial sector, health, infrastructure, labor and social protection, private sector and trade, and public sector (see Table 3).

Methods
Based on an initial set of unclassified data, cluster analysis constructs a sensible and informative classification using the values of the observed variables (Everitt 2002). Single linkage clustering was used in the present analysis. The distance between two clusters was defined as the minimum distance between a pair of individuals (Everitt 2002). As explained by Larose (2005), clustering refers to the grouping of records, observations, or cases into classes of similar objects and clustering differs from classification in that there is no target variable for clustering.
As pointed out by Larose (2005), clustering algorithms seek to segment the entire dataset into relatively homoge neous subgroups or clusters, in which the similarity of the records within the cluster is maximized and the similarity to records outside this cluster is minimized (see Figures 3 and 4). Johnson (1967) stated that hierarchical clustering starts with a set of N items to be clustered and an N × N distance (or similarity) matrix. In an iterative procedure, as depicted in Figure 5, it searches for the closest (most similar) pair of clusters and merges it into a single cluster. Then, it computes the distances (similarities) between the new cluster and each of the old clusters to update the similarity matrix. These two steps are repeated until all items are clustered into a single group of size N. In order to generate k clusters, the algorithm can be stopped after N k iterations (Dressler 2008).
In each iteration, the minimum distances between the members of all established clusters are computed and the two closest clusters are merged accordingly. The algorithm terminates if the whole dataset is merged into a single clus ter (Dressler 2008). For optimal performance, clustering algorithms, just like algorithms for classification, require the data to be normalized so that no particular variable or subset of variables dominates the analysis (Larose 2005). Analysts may use either the min-max normalization or Zscore standardization (Larose 2005).

Clustering analysis for Turkey and the EU
In this section, the targeted number of groups is deter mined by evaluating the data and distances between va rious clusters. The optimal number of targeted clusters is chosen by taking into account the distances between clusters using agglomeration plots. In addition, cluster group numbers are sorted and values placed into ascen ding or descending order. For instance, in Table 4, Turkey ranks number 1, which is cluster group 1 for the Env4 variable. This implies that in addition to sharing its group with Romania, the electric power consumption per capi ta (Env4) of Turkey is lower than EU members. For the Env11 Livestock production index, Turkey appears in the  1. Assign each item to a cluster (N items result in N clusters each con taining one item) 2. Let the distances between the clusters be the same as the distances between the items they contain 3. repeat 4. Find the closest pair of clusters and merge them into a single cluster 5. Compute distances between the new cluster and each of the old clusters 6. until All items are clustered into a single cluster of size N in the last cluster, their livestock production indexes are higher than other member countries.
Further, as shown in Table 4, for the Env3 variable, Turkey belongs to the second of the five clusters, which is represented by the majority of European countries (85.7%). This implies that Turkey is not significantly different from the listed countries in this regard. For Env1, Env4, and Env11, Turkey shares its cluster with Poland, Romania, and Latvia.
For Env10, an important environmental indicator, Turkey is in the third cluster of the seven clusters in total. Although there exists a strong tendency in Turkey towards reducing fossil fuel usage, there is no simple short cut for eliminating its use. It is well known that fossil fuels have limited potential and, at the current rate of exploitation, they are expected to be depleted within the next few centu ries. However, despite its eastern and southern neighboring countries, Turkey has limited fuel oil reserves, and most of its highquality coal reserves have already been used (Sahin 2008). By contrast, Turkey is a leading country for wind energy potential, one of the renewable energy sources being promoted to reduce the amount of carbon dioxide in the atmosphere (Sahin 2008). It is expected that this source will be one of the main driving forces for future industrial development in Turkey (Sahin 2008).
According to Econ1 in Table 5, Turkey is placed in the last group, as it has the highest percentage of GDP value added for agriculture among the 28 countries. Further, for Econ 2, Econ11, and Econ15, Turkey has the highest ran ked cluster number. Finally, Turkey's percentage of GDP is low along with the countries within its cluster in exports of goods and services (Econ2), imports of goods and services (Econ11), and trade (Econ15) in comparison with other member countries.
By taking into consideration Turkey's essential econo mic policy and debt indicators, Kizilaslan et al. (2007) re ported that Turkish gross national income seems to be in the range of critical development based on the country's population density and forest areas. Moreover, in terms of its development, agricultural selfsufficiency rate, and urban population being in the range of sustainable development, Turkey is expected to produce effective policies for hazar dous and critical values (Kizilaslan et al. 2007). Kavalsky (2006) found that considerably reducing agricultural subsi dies and reducing the ability to fund offbudget subsidies by public banks contribute to sustainable fiscal improvement. However, the deficit in the pension system has grown rapi dly, offsetting much of this gain (Kavalsky 2006). Moreover, Kavalsky's (2006) report to the World Bank further showed that agricultural reforms have reduced and that rationalized subsidies and price supports have replaced state marketing agencies with private commodity exchanges, thereby redu cing food costs for consumers.
For Fin3, Fin7, Fin8, Fin11, and Fin12, Turkey tops the grouping numbers in cluster number 1, whereas for   Fin4, Fin5, and Fin10, Turkey has significant dissimila rities among the clustered countries (Table 6). In Turkey, the financial sector reforms in the early 1980s were un successful; however, by the late 1990s, the banks and go vernment agreed on a strategy to reform the sector (Effron 2006). These strategies created strong agency regulation and supervision for banks, aligned prudential regulations with international standards, strengthened the agency failed bank resolution (entity deposit insurance), and restructured and privatized public banks (Effron 2006). Kavalsky (2006) also reported that private investment and a surge in exports following trade liberalization helped Turkey grow rapidly during the 1980s. However, a mixture of public investment in infrastructure and populist policies such as generous pensions for civil servants and large agricultural subsidies gradually led to an imbalance in public accounts and high inflation (Kavalsky 2006). In Turkey, following significant growth and a rising primary surplus after 2001 as well as three volatile financial crises, extra budgetary funds were eliminated and fiscal controls streamlined (Kavalsky 2006).  In addition, the Banking Regulation and Supervisory Authority established an independent regulatory and su pervisory framework that was closely aligned with EU stan dards (Kavalsky 2006). Although this measure strengthened the banking system, private sector credit as a percentage of GDP remained low compared with the OECD average and little progress was made towards the privatization of public banks (Kavalsky 2006).
As shown in Table 7 for Hlth4 and Hlth5, Turkey's group is represented by 68.0% and 42.9% of the countries, respectively. For Hlth6 and Hlth1, Turkey is placed in the first cluster; indeed, Turkey's death rate (Hlth1) is the lowest among member countries, while it has the lowest number of physicians per 1,000 people (Hlth6) among members.
YenimahalleliYasar and Ugurluoglu (2011) explained that the conditions for receiving bonuses for health services and the debt premium would continue to shrink actual he alth coverage, while outofpocket payments and expenses would continue to affect access to services. In particular, YenimahalleliYasar (2011) pointed out that meeting health objectives seems to take a long time, as many factors, both related and unrelated to health policy, are responsible for the low level of health improvements in Turkey, whose GDP per capita and education level (especially among women) is low. Thus, additional investment in the prevention of social policy issues and other health matters is an important is sue (YenimahalleliYasar 2010). In addition, Yildirim et al. (2011) reported shortages in medical staff, the unequal distribution of resources across the nation, and the lack of systematic information on results, thereby limiting the selection of providers for many people.
For Infra4, Turkey shares the top cluster with 21 mem bers (81.5%), whereas for Infra1, Turkey has the lowest rank (Table 8). Kavalsky (2006) provided a general summary of how Turkey's regulatory frameworks and institutions have been established and energy, telecommunications, and railways improved. Following the EU approach to libera lization, Turkey liberalized its telecommunications sector. However, there remain huge opportunities for Turkey to benefit from the adoption and implementation of a le gislative, regulatory, and institutional framework in the telecommunications sector (Akdemir et al. 2007). In addi tion, according to Kavalsky (2006), private investment is growing in power and public services, but quantifiable gains in effectiveness have not yet emerged. From Table 9, the Lbr1 and Lbr15 indicators show that Turkey has the highest rate of contributing family workers but also the highest group of vulnerable employees among member states. In addition, Turkey is in the first cluster (i.e., lowest rates) for Lbr7, Lbr8, and Lbr16. For Lbr7, Turkey shares a cluster with Romania, Portugal, Czech Republic, Italy, Slovak Republic, and Malta. Hoekman and Togan (2005) stated that Turkey's age composition is one of its most important characteristics; owing to a high birth rate, the population is relatively young. This high proportion of young people could be an advantage for Turkey, because it includes a significant proportion of the workforce, but it imposes a heavy burden on the education system and employment generation is a major social issue (Hoekman, Togan 2005). Additionally, the share of emplo yment in agriculture in Turkey is extremely high (34.5%) among candidate countries. Most selfemployed and part time employees work in agriculture, and fixedterm emplo yment is dominant in the construction sector (Hoekman, Togan 2005). Its possible accession to the EU would thus have a profound impact on Turkey and EU countries, and this impact would be largely determined by the peculiarities of the structure of the population and Turkish labor markets (Hoekman, Togan 2005).
As shown in Table 10, Prv7 and Prv12 are indicators in which Turkey ranks in the first cluster. Under the Prv7 indicator, Turkey shares its cluster with Hungary; these two countries show the lowest rates of food imports among ot her member countries. Under the Prv12 indicator, Romania, Czech Republic, Slovak Republic, Lithuania, Hungary, and Greece are in the same cluster; these nations have the lowest rate (as a percentage of total imports) among other members.
The transformation of the Turkish economy and its cur rent trade patterns are similar to those of other emerging market economies. After the adoption of a business strategy directed outwards in the 1980s, Turkey was able to increase its share of world exports from 0.36% in 1980 to 0.60% in 2000. Meanwhile, the average growth of exports exceeded 20% after the financial crises in 2001 (Saygili, Saygili 2011). The Turkish economy features a relatively large share of value added in agriculture (14.2%). This share is smaller than those for Bulgaria and Romania, where the agricul tural sector comprises 28.2% and 19.3% of total value ad ded, respectively (Lejour, de Mooij 2005). It is much larger, however, than in the Accession10, where the agricultural sector is responsible for 6.9% of value added on average, and the EU15 where it is only 2.5%. One reason for the large agricultural sector in Turkey is the substantial amount of agricultural support by the Turkish government (Lejour, de Mooij 2005). The trade liberalization in the 1980s and 1990s, which allowed Turkey to expand business contacts with glo bal production networks, resulted in a transfer of resources from traditional sectors such as textiles and agriculture to nontraditional high technologyintensive sectors such as transport vehicles and consumer electronics (Saygili, Saygili 2011). Trade liberalization has thus been an important as pect of economic policy in Turkey since the early 1980s. It also led to the formation of the customs union between Turkey and the EU in 1995, which not only covers trade in industrial goods and processed agricultural products but also covers the harmonization of technical legislation, the abolition of monopolies, and the protection of intellectual property (Lejour, de Mooij 2005).
Apart from agriculture, Turkey also has relatively lar ge textiles, trade services, and transport services sectors. These sectors are laborintensive and have relatively low productivity levels. Further, the tourism sector is impor tant for the Turkish economy. However, compared with the Accession10, Turkey has a low share in the machinery and equipment, transport equipment, and business services sectors (Lejour, de Mooij 2005).
For Pblc4, Pblc10, and Pblc11, Turkey stands in the first cluster with the lowest rates of these indicators among other members. For the Pblc1 indicator, Turkey is in the sixth cluster alongside Latvia, Czech Republic, Poland, Cyprus, Slovenia, Belgium, Italy, the Netherlands, Hungary, and Malta (see Table 11). Kavalsky (2006) reported that the private sector in Turkey requires a substantial increase in foreign direct investment to be more competitive and technologically advanced (Basti, Bayyurt 2008). For Turkey, increasing R&D, improving intellectual property rights, and strengt hening the institutional framework for technical standards and finance are necessary to gain in selfsufficiency. In addi tion, progress in the management of public procurement, fiscal transparency, and energy and banking regulation should reduce corruption. The poverty rate is also declining based on the trend in GDP per capita growth, and poverty monitoring is improving. However, there is a lower employ ment rate for women compared with men; employment, in particular, has been very slow. Moreover, the financial terms of the retirement system is considered to be unsustainable, although the regional distribution of income does not seem to be damaged. In addition, there has been a sharp decline in infant mortality (45%) and a rapid increase in school enrollment (98% gross of 84%), especially girls at secondary level. Literacy has also developed for both men and women.

Conclusion
This paper reviewed the path of Turkey's accession efforts to the EU and used cluster analysis to determine the signifi cance of trade ties and the country's possible standing com pared with other nations. The cluster analysis technique was used to measure Turkey's similarities with other member countries. Recent data on the economic and development indicators of countries from the World Bank was used for the cluster analysis. From the World Bank database, indicators on the envi ronment, economic policy and debt, financial sector, health, infrastructure, labor and social protection, private sector and trade, and public sector were taken into consideration. To form the clusters, the procedure began with each obser vation in a separate group. It then combined the two closest observations in order to form a new group. After recalcula ting the distance between the groups, the two closest were combined. This process was repeated until only the targeted number of groups remained.
Turkey is playing an increasingly influential role in the region because of not only its growing economic power but also its increased strategic importance. Further, with its stable government and strategic location, Turkey is beco ming a crucial partner of the west in terms of helping solve many globalscale conflicts and problems. From an econo mic point of view, Turkey has seen promising growth and is aiming for the highest standards of living with a constantly increasing purchasing power in years to come. Indeed, while Turkey is still negotiating to join the EU, the eurozone is struggling with sovereign debt.

APPENDIx A
The output summarizes the statistical significance of the terms in the GDP forecasting model. Terms with Pvalues less than 0.05 are statistically significantly different from zero at the 95% confidence level. The Pvalue for the MA(1) term is less than 0.05, so it is significantly different from 0. The estimated standard deviation of the input white noise equals 1288.77. Each of the statistics is based on the oneahead forecast errors, which are the differences between the data value at time t and the forecast of that value made at time t1. The first three statistics measure the magnitude of the errors. A better model will give a smaller value. The last two statistics measure bias. A better model will give a value close to 0. Table A3 shows the forecasted values for GDP per capita. During the period where actual data is available, it also dis plays the predicted values from the fitted model and the residuals (dataforecast).  Table A4 compares the results of fitting different models to the data. The model with the lowest value of the Akaike Information Criterion (AIC) is model J, which has been used to generate the forecasts.