Öncül göstergeler yöntemi ve Türkiye uygulaması
Leading indicators and an application on Turkish data
- Tez No: 46377
- Danışmanlar: PROF.DR. ÜMİT ŞENESEN
- Tez Türü: Doktora
- Konular: Mühendislik Bilimleri, İşletme, Engineering Sciences, Business Administration
- Anahtar Kelimeler: Belirtilmemiş.
- Yıl: 1995
- Dil: Türkçe
- Üniversite: İstanbul Teknik Üniversitesi
- Enstitü: Fen Bilimleri Enstitüsü
- Ana Bilim Dalı: Belirtilmemiş.
- Bilim Dalı: Belirtilmemiş.
- Sayfa Sayısı: 72
Özet
ÖZET Ekonomik olayların zaman içerisinde aldığı değerleri gösteren zaman serileri incelendiğinde bazı dalgalanmalar (iniş ve çıkışlar) ve düzensizlikler görülmektedir. Ekonomik yaşamla bağlantılı kararlarda, karar vericiler için, ekonomik göstergelerdeki bu tip dalgalanmaların değerlendirilmesi son derece önem taşımaktadır. Ekonomik yaşamda yön değiştirmelere neden olan serilerin devresel hareketlerinin belirlenmesi ve bu hareketlerden yararlanılarak yapılacak geleceğin tahmini, kısa dönem ekonomi politikası ile ilgili kararlar bakımından büyük yararlar sağlamaktadır. Ekonomide görülen devresel dalgalanmaların belirlenmesini temel alan öncül göstergeler yöntemi, toplumun (insanların) daha önceki davranışlarına dayanan ve yeni hükümet politakalarının ekonomik davranışları değiştirmesinden olumsuz yönde etkilenen tahmin yöntemlerine alternatif bir yöntem olarak kullanılmaktadır. Bu çalışma, öncül göstergeler yöntemini açıklamak ve Türkiye ekonomisi için genel ekonomik durumun önceden tahminini olanaklı kılacak gösterge özelliklerine sahip serileri kapsayacak bir öncül gösterge indeksi oluşturmak amacıyla yapılmışıtır. 1978-1993 dönemini kapsayan bu çalışmada, ekonominin kısa dönemli analizini yapabilmek için Türkiye ekonomisini yansıtan 9 ana grupta, aylık yayımlanan 84 seri incelenmiştir. Yapılan sınamalar sonucunda ekonomik durumu en iyi yansıtacak referans seri olarak sanayi üretim indeksi seçilmiş ve ekonomik durum indeksi hesaplanmıştır. Tüm serilerin referans seriyle olan uyumluluğunun tüm çevrim ve dönme noktaları yaklaşımlarına göre incelenmeleri sonucunda ise, genel ekonomik durumdaki gelişmelerden önce gelen 6 öncül serinin öncül göstergeler indeksine girebileceğine karar verilmiştir. Altı aday öncül serinin (iş isteyen sayısı, ham petrol üretimi, sanayi yapı kullanımı, reel para arzı (M2), işçi gelirleri ve ulusal net rezerv) bileşik indeks yöntemine dayandırılarak incelenmesi sonucunda öncül göstergeler indeksinde, iş isteyen sayısı, reel para arzı (M2) ve sanayi yapı kullanımı serilerinin kullanılması uygun bulunmuştur. Çalışmadan elde edilen Türkiye ekonomisi için oluşturulan öncül göstergeler indekslerinin ekonomik durum indeksine göre ortalama 7.4 - 9.3 ay önden gitmesi bulgusundan hareketle, bu indekslerin düzenli bir biçimde hazırlanmasının, ekonomik durumun 7-9 ay sonrası hakkında güvenilir tahminlerin yapılmasına yardımcı olacağı sonucuna varılmıştır. vm
Özet (Çeviri)
SUMMARY LEADING INDICATORS AND AN APPLICATION ON TURKISH DATA The main purpose of this study is to construct an Index of Leading Indicators for the Turkish economy. This index is made of the series that conform the leading indicators' criteria and have high correlation with the economic activity. The first section of this study is the introductory section and explains the main theme of the thesis. The purpose of the study and the time limitations of the subject are also explained in this section. In the second section, the scientific forecasting methods are explained and the position of the leading indicators methods among the forecasting methods is examined. Advantages and disadvantages of this method from similar methods are underlined. The third section comprises the theoretical basis of the thesis. In this section, a brief introduction to the classical time series model is given and its basic components are pointed out. Then, procedures and the methodology for forecasting the classical component of a time series are given. The cycle, upswing, downswing, peak, trough, recovery, prosperity, recession and depression concepts are defined. The fourth section begins with the application steps of the method. Then,“reference series”, coincident, lead and lag indicators are pointed out. In addition, two approaches used in evaluating the cyclical conformity of the candidate leading indicator to the indicator indexes are explained. These are called as the whole- cycle approach and the turning point approach. This section finishes with the construction and the formulation of the composite indicator indexes. The fifth section includes the history of leading indicators method. In this section, after explaining the position of cyclical movements in the theory of economy, information about applications and series used in the United States and other developed European countries is given. Finally, previous works on leading indicators for Turkey are evaluated. The sixth section of this study involves the application carried out by us for the Turkish economy. In the first part of this section, analysed series and the IXperiod of the practice in the work are mentioned. The second part explaines the choice of reference series and the structure of“Industrial Production Index”which is considered as a reference series. This part finishes with the detection of cyclical movements and the identification of turning points. The study continues with the clasification of the candidate series into coincident with and leading the indicator index. The sixth section ends by comparing the leading indicators index with the one for the economic activity. In the final section, the general findings of the study are discussed. In the appendices, there are references and the data bank of all series used. When the forecasting techniques classification is examined, approximately fifty different scientific forecasting techniques can be seen (Chambers, Mullick and Smith 1971 and 1974, Makridakis and Wheelwright 1978 ve 1983, Armstrong 1985, Georgoff and Murdick 1986 and Simmons and Wright 1990). The method of leading indicators, which is one of these techniques, is based on the cyclical movements of the economic series. This method is an example of an approach to forecasting which represents a major alternative to econometric models. Construction of the index does not rely on type of the theory embodied in econometric models. The use and construction of the leading indicators are both easy. The leading indicator approach to forecasting does not require assumptions about the causes of people's economic behavior. Instead, it relies on statistically detecting patterns among economic variables which can be used to forecast turning points in economic activity (Chambers and others, 1971, p.45-74, Makridakis and Wheelwright, 1978, p.440- 443, Georgeff and Murdick 1986, p. 110-120, Gorton, 1982, p.18, Ferman, 1988, p.57-62, Özmucur, 1990, p. 15-1 6). The use of forecasting of leading economic indicators in business cycle -as originally developed by Wesley Mitchell and Arthur Burns (1 920's) at the National Bureau of Economic Research- has continued and flourished in popularity. This forecasting approach is explained by the simple logic: if the indicator goes one way today economic activity will go the same way tomorrow. According to the classical time series model, an economic series (Y) is defined either by the multiplicative components or the additive components [(Y=T*M*C*D) or (Y=T+M+C+D)] as follows. (i) Trend (T) (ii) The seasonal movements (M) (iii) The cyclical movements (C) (iv) The random movements (D) (Farnum and Stanton, 1989, p.409, Makridakis and Wheelwright, 1978, p. 88- 90, Sanders, 1990, p.509-510).A cycle from the cyclical movement can be seen as below. The Phases of a Cycle (Akgür, 1990, p.22). The cyclical movement components of the candidate series to the indicator indexes can be measured by its separation from other components of the economic series. A candidate indicator series is first deseasonalised by using the“X11 -Census”method and then the trend of the series is calculated by using the“ordinary least squares”method (Cleveland and Devlin, 1980, p.487, Cleveland and Tiao, 1976, p.581, Makridakis and Wheelwright, 1978, p.88-138). The moving average values (=T*C) are divided by the calculated trend values to obtain the cyclical movement component. Then the standardized cyclical movement series are indexed as reference series and the leading indicator series. The standardized series are indexed on the basis of the whole cycle approach and the turning point approach. The whole cycle approach. This method have been used to investigate the relations between the index of leading indicators and economic activity over all points in the cycle rather than just near turning points. Thus, certain movements in one series, which are associated with movements in another series, can be identified. The cross-correlation coefficient is a tool for the whole cycle approach. The cross-correlation coefficient is 'XT, (/+*> i XIA- = -î, 7 = ^ - k =, -3, -2, -1, 0, 1, 2, 3,..(the number of months). X = The standardized series, Y = The standardized reference series. The standard error of the cross-correlation coefficient is, 1 Jn - \k\ n = the number of observations Hi.. P^ * ° (Farnum ve Stanton, 1989, p.427-428) The turning point approach. This method is based on the peak and trough analysis. First the graph of standardized cyclical movement of series is drawn and then the peak and trough points of movement are determined (Hymans, 1973, p.339-341, Lesage, 1992, p.37, Niemara and Fredman, 1991, p.49-50, Stekler and Schepsman, 1973, p.291). Cyclical indicators are evaluated on the basis of six major characteristics (Auerbach, 1982, p. 590, Chisholm and Whitaker, 1971, p. 47-48, Gorton, 1982, p. 23-24, Neftçi and Özmucur; 1991, p.6, Renshaw 1987, p.627, Withycombe, 1982, p.19). Economic significance. The indicator must measure or represent an activity with a key role in the cyclical process. Statistical adequacy. The indicator must be based on well-established, accurate reporting systems. Timing. The indicator must have the same pattern of growth and contraction as the reference series. To be classed as a leading indicator, its cycle must appear earlier. An indicator is of no use as a leader if it regularly declines and recovers after the overall economy has already done so. This factor is considered the most important of these six selection criteria. XUConformity. The indicator's decline and recovery pattern should conform closely to past business cycles as measured as changes in reference series. Smoothness. A smooth indicator (one that has few irregular changes) is more likely to give prompt notice as to when a change has occured, thus making it more valuable as a predictor. Prompt availability. To be useful, information about the indicator must be available on a timely basis. If current data for an indicator are not available then it obviously will be of little use as an indicator. The series are chosen according to these criteria are applied equal or unequal weights and the indicator index can be calculated as in the following equation. m wj = the weight of the cyclical movement j Cjt = the value of the cyclical movement j at the period t It = the value of index at the period t t = the period number m = the number of indicators (Kozlowski, 1987, p.64, Martin, 1990, p.658, Neftçi and Özmucur, 1991, p. 13). The application carried out by us for Turkish economy considered the period between 1978 and 1993. In this study, the 84 series published monthly in Turkey were examined and the industrial production index was used as“reference series”as it is with most O.E.C.D. contries. The candidate leading series were chosen according to cyclical conformity between each of the series and the reference series. These series are the number of application for employment, the production of crude oil, the building permits for industrial construction, the real money supply (M2), the workers' remittances and the national net reserves. At the end of the analysis, six candidate leading series were observed and three different leading indicator indexes were calculated. The final leading indicator indexes was composed of the applications for employment, the building permits for industrial construction and the real money supply (CIIS-CIRS, CIIS- CREM2, CIIS-CIRS-CREM2). xmAccording to the results of our study, the composite leading indicator indexes lead the economic activity (reference series) index by 7.4-9.3 months on avarage. The highest cross-correlation between the composite indexes and the reference series for the nine months after the estimation period is 0.835. If these leading indicator indexes are calculated regularly, the reliable predictions for the future (7-9 months) economic activity can be obtained. XIV
Benzer Tezler
- Finansal krizler ve erken uyarı sistemleri: Türkiye uygulaması (1988-2012)
Financial crises and early warning systems: An application to Turkey (1988-2012)
FULYA SEYREK
- Ekonomide öncü göstergeler ile büyüme tahmini uygulaması
An application of growth prediction in economy with leading indicators
EZGİ DÜLGER
Yüksek Lisans
Türkçe
2016
Bilgisayar Mühendisliği Bilimleri-Bilgisayar ve KontrolAnkara ÜniversitesiBilgisayar Mühendisliği Ana Bilim Dalı
YRD. DOÇ. DR. MEHMET SERDAR GÜZEL
- Ekonomik krizlerin öncü göstergeleri: KLR modeli Türkiye uygulaması
Early warning indicators of economic crises: KLR model application of Turkey
ATİLLA YILDIZ
Yüksek Lisans
Türkçe
2018
EkonomiSakarya Üniversitesiİktisat Ana Bilim Dalı
DR. ÖĞR. ÜYESİ ŞÜKRÜ CİCİOĞLU
- Hayatdışı sigorta şirketlerinde mali yeterliliğin ölçülmesi ve Türkiye için bir erken uyarı modeli önerisi
Solvence measurement of non-life insurance companies and an early warning model proposal for Turkey
AHMET GENÇ
- A comparative evaluation study on intellectual property valuation methods
Fikri sınai mülkiyet hakları değerleme yöntemlerinin karşılaştırmalı değerlendirilmesi üzerine bir çalışma
AYSUN BEYAZKILIÇ KOÇ
Yüksek Lisans
İngilizce
2018
Bilim ve Teknolojiİstanbul Teknik Üniversitesiİşletme Mühendisliği Ana Bilim Dalı
DOÇ. DR. NİHAN YILDIRIM