【摘要】：市場經(jīng)濟所表現(xiàn)出來的擴張與收縮交替出現(xiàn)的經(jīng)濟周期現(xiàn)象,一直是宏觀經(jīng)濟學(xué)關(guān)注的焦點。自凱恩斯以來,相繼出現(xiàn)的每個宏觀經(jīng)濟學(xué)派都在試圖創(chuàng)立新的理論來解釋經(jīng)濟周期,如何對這些競爭性的理論進(jìn)行評價,便成為宏觀經(jīng)濟學(xué)研究的主要內(nèi)容。以宏觀經(jīng)濟統(tǒng)計數(shù)據(jù)為基礎(chǔ)的經(jīng)驗研究,是對經(jīng)濟周期理論進(jìn)行評價的主流方法,而進(jìn)行這些經(jīng)驗研究的前提是能夠提取宏觀經(jīng)濟數(shù)據(jù)中的周期成分,這正是本文所關(guān)注的,即趨勢周期分解方法。2004年以來,國內(nèi)對這一領(lǐng)域的研究也逐漸增多,但仍然處在對各種分解方法進(jìn)行應(yīng)用的階段。 本文根據(jù)方法論的不同將趨勢周期分解方法分成單變量分解方法和多變量分解方法,其中前者還可以細(xì)分為基于概率理論的方法和濾波方法,本文的結(jié)構(gòu)也據(jù)此安排,現(xiàn)將本文研究的主要內(nèi)容及創(chuàng)新之處總結(jié)如下。 在單變量趨勢周期分解領(lǐng)域,本文具有理論創(chuàng)新意義的研究是,針對目前仍然沒有得到徹底解決的BN分解周期成分的“反號”問題,從方法論的角度進(jìn)行了研究,建立了此現(xiàn)象與持久性度量指標(biāo)之間的定量關(guān)系。研究的結(jié)論為：當(dāng)以方差比為代表的持久性度量指標(biāo)大于1時,表明序列存在很強的持久性,周期成分的“反號”問題便會出現(xiàn),否則不會出現(xiàn)。這一結(jié)果顯示了比較清晰的理論創(chuàng)新意義。本文還發(fā)現(xiàn)不可觀測成分模型分解的周期成分同樣存在著“反號”問題,并且同樣可以建立類似的定量關(guān)系。以上兩方面的理論研究,在趨勢周期分解領(lǐng)域尚屬首次。 在解析單變量分解的方法論的基礎(chǔ)上,本文針對分解的周期成分,提出周期性、穩(wěn)定性、及時性和一致性等標(biāo)準(zhǔn),對各種分解方法進(jìn)行評價。周期性關(guān)注的是,分解的周期成分的波動周期是否與需要考慮的經(jīng)濟周期相一致。穩(wěn)定性關(guān)注的是,加入新的觀測值后,最新分解的周期成分偏離之前的周期成分的程度。及時性關(guān)注的是,分解的周期成分的末期值能否準(zhǔn)確地反映此變量的現(xiàn)狀。一致性關(guān)注的是,對于不同頻率的數(shù)據(jù),該分解方法是否能夠得到相似的周期成分。針對我國實際GDP變量的研究顯示,CF濾波相對比較適合提取其周期成分。從比較研究的角度來看,本文提出的標(biāo)準(zhǔn),能夠較好地對各種分解方法進(jìn)行評價,從而為實證分解類似于我國GDP這樣的宏觀數(shù)據(jù)提供理論基礎(chǔ),由此體現(xiàn)了理論和應(yīng)用的創(chuàng)新意義。 在趨勢周期分解領(lǐng)域,多變量分解是最為前沿也是最為困難的研究方向。本文跟蹤主要的研究文獻(xiàn),從方法論的角度,將其發(fā)展脈絡(luò)和分解理論進(jìn)行了清晰地解讀和表述。從文獻(xiàn)看,到目前為止,國內(nèi)還沒有關(guān)于這一領(lǐng)域的相關(guān)研究工作。具體而言,多變量趨勢周期分解是以非平穩(wěn)時間序列理論為基礎(chǔ),在考慮了各趨勢之間的相互影響和各周期之間的相互影響后,對多個變量同時進(jìn)行分解的方法。為此,本文首先通過協(xié)整理論對多個趨勢之間的相互影響進(jìn)行研究,再通過Engle和Kozicki(1993)提出的序列相關(guān)共同特征理論對多個周期之間的相互影響進(jìn)行研究。在此基礎(chǔ)上,通過包含協(xié)整和序列相關(guān)共同特征約束的VEC模型來實現(xiàn)對多變量的趨勢周期分解。在多變量分解方法論研究的基礎(chǔ)上,本文對我國現(xiàn)實經(jīng)濟問題進(jìn)行了應(yīng)用研究。針對貨幣與經(jīng)濟運行和價格水平的研究顯示,貨幣因素在經(jīng)濟周期過程中扮演著重要的角色,貨幣供給量是經(jīng)濟運行的先行指標(biāo)。針對農(nóng)產(chǎn)品價格與通貨膨脹問題的研究顯示,農(nóng)產(chǎn)品價格周期與CPI周期的相依性較弱,農(nóng)產(chǎn)品價格周期所呈現(xiàn)的大幅波動,在很大程度上為糧食、豬肉價格的暴漲暴跌等因素的沖擊效應(yīng)。貨幣周期與CPI周期的相依性較強,說明我國貨幣政策目標(biāo)在抑制通脹和促進(jìn)增長之間交替轉(zhuǎn)換。據(jù)此,本文還給出在農(nóng)產(chǎn)品價格周期的上升期,調(diào)控通脹應(yīng)以從供給方面抑制農(nóng)產(chǎn)品價格為先導(dǎo)等政策建議。 綜上,本文理論和方法論的創(chuàng)新之處可以概括為：對BN分解和UC模型分解方法的周期成分的“反號”問題的研究,通過建立評價標(biāo)準(zhǔn)對各種單變量分解方法進(jìn)行評價的研究。應(yīng)用性創(chuàng)新體現(xiàn)在,利用新穎的多變量趨勢周期分解這種計量經(jīng)濟學(xué)方法,從重要但還沒有得到充分重視的角度出發(fā),研究我國經(jīng)濟運行和通貨膨脹等問題,得出具有現(xiàn)實意義的結(jié)論和具有應(yīng)用價值的政策建議。
[Abstract]:The economic cycle phenomenon appearing alternately between expansion and contraction in the market economy has always been the focus of macroeconomics. Since Keynes, every macroeconomic school that has appeared successively is trying to create new theories to explain the economic cycle. How to evaluate these competing theories has become macro economics. The main content of the study. The empirical research based on the macroeconomic statistics is the main method to evaluate the economic cycle theory, and the premise of these empirical studies is to extract the periodic components in the macro-economic data. This is the focus of this article, that is, the trend cycle decomposition method in.2004 years. Research in the field has gradually increased, but it is still in the stage of application of various decomposition methods.
This paper divides the trend periodic decomposition method into the single variable decomposition method and the multivariable decomposition method according to the different methodology, and the former can also be subdivided into the method based on probability theory and the filtering method. The structure of this paper is also arranged accordingly. The main contents and innovations of this paper are summarized as follows.
In the field of univariate trend periodic decomposition, this paper has the significance of theoretical innovation. In view of the "anti sign" problem of the periodic component of BN decomposition, which has not been thoroughly solved at present, the research is carried out from the perspective of methodology, and the quantitative relationship between this phenomenon and the persistence metric is established. The persistence of the difference ratio is more than 1, which indicates that the sequence has a strong persistence, and the "anti sign" problem of the periodic component will appear, otherwise it will not appear. This result shows a relatively clear theoretical innovation. This paper also finds that the periodic component of the unobservable component model also has the "anti sign" question. The above two aspects of theoretical research are the first time in the trend cycle decomposition field.
On the basis of analyzing the methodology of single variable decomposition, this paper aims at the periodic, stability, timeliness and consistency criteria of the decomposition of the decomposition, and evaluates the various decomposition methods. The periodic concern is whether the periodic component of the decomposition is in accordance with the economic cycle that needs to be considered. After adding new observation values, the degree of the periodic component of the latest decomposition deviates from the previous periodic component. The time is concerned that the end value of the periodic component of the decomposition can accurately reflect the status of the variable. The study of the actual GDP variables in the country shows that the CF filter is relatively suitable for extracting its periodic components. From the perspective of comparative study, the standards proposed in this paper can better evaluate various decomposition methods, thus the empirical decomposition is similar to the theoretical basis of the macro data provision, which is similar to our country's GDP, thus reflecting the creation of theory and application. New meaning.
In the field of trend periodic decomposition, multivariable decomposition is the most frontier and the most difficult research direction. This paper traces the main research literature and clearly interprets and expresses its development and decomposition theory from the perspective of methodology. From the literature, so far, there is no relevant research work in this field at home. In particular, the multivariable trend periodic decomposition is based on the non stationary time series theory. After considering the interaction between the various trends and the interaction between each cycle, the multivariable decomposition method is taken into account. The theory of sequence related common feature proposed by Engle and Kozicki (1993) is used to study the interaction between multiple cycles. On this basis, the trend periodic decomposition of multivariables is realized by the VEC model containing cointegration and sequence related common characteristic constraints. On the basis of the study of the multivariable decomposition methodology, this paper is present to our country. The study of real economic problems is applied. According to the study of the monetary and economic operation and the price level, the monetary factor plays an important role in the economic cycle. The supply of money is the first indicator of the economic operation. The research on the price and inflation of agricultural products shows the phase of the price cycle of agricultural products and the CPI cycle. The large fluctuations in the price cycle of agricultural products, to a large extent, are the impact of factors such as the rise and fall of the price of pork, and the strong dependence of the monetary cycle with the CPI cycle shows that the monetary policy goal of our country is alternately converted between inflation and growth. Accordingly, the price of the agricultural product is also given in the price of agricultural products. In the rising period of the grid cycle, inflation control should be guided by the policy of restraining the prices of agricultural products from the supply side.
To sum up, the innovation of the theory and methodology of this paper can be summarized as the study of the "anti sign" problem of the periodic components of the BN decomposition and the UC model decomposition method, and the evaluation of various single variable decomposition methods through the establishment of evaluation criteria. The application innovation is embodied in the new multivariable trend periodic decomposition of this measure. The economic method, from the point of view of important but not fully paid attention to, studies the problems of economic operation and inflation in China, and draws realistic conclusions and policy suggestions with practical value.
【學(xué)位授予單位】：華中科技大學(xué)
【學(xué)位級別】：博士
【學(xué)位授予年份】：2013
【分類號】：F224;F124.8

【相似文獻(xiàn)】

相關(guān)期刊論文 前10條

1 李興緒;李時興;;能源消費與經(jīng)濟增長關(guān)聯(lián)特征的分解分析[J];云南財經(jīng)大學(xué)學(xué)報;2008年05期

2 王明利;李威夷;;生豬價格的趨勢周期分解和隨機沖擊效應(yīng)測定[J];農(nóng)業(yè)技術(shù)經(jīng)濟;2010年12期

3 朱玉成;滬幫裁縫[J];上海工藝美術(shù);1997年01期

4 陳穎濤;國有企業(yè)經(jīng)營趨勢周期的實證分析[J];數(shù)量經(jīng)濟技術(shù)經(jīng)濟研究;2002年02期

5 劉金全,劉志剛;我國GDP增長率序列中趨勢成分和周期成分的分解[J];數(shù)量經(jīng)濟技術(shù)經(jīng)濟研究;2004年05期

6 劉東紅;;河南省經(jīng)濟增長的趨勢變化與波動分析[J];經(jīng)濟研究導(dǎo)刊;2011年29期

7 周建;;基于隨機方差擴大模型的對中國宏觀經(jīng)濟統(tǒng)計數(shù)據(jù)的結(jié)構(gòu)變化分析[J];中國管理科學(xué);2006年03期

8 謝富華;徐莉莉;;底部探明 買入股票型基金[J];證券導(dǎo)刊;2009年42期

9 唐麗佳;;我國房地產(chǎn)市場的價格動態(tài)研究[J];商業(yè)文化(上半月);2011年04期

10 許劉俊,張漢昌,梁志榮,廖志芳,陳潔玲;一種測度廣東經(jīng)濟周期波動的模型[J];統(tǒng)計與預(yù)測;1996年06期

相關(guān)會議論文 前7條

1 鄧澄;;徑流的一致性分析——以新疆額爾齊斯河為例分析研究[A];中國水力發(fā)電工程學(xué)會水文泥沙專業(yè)委員會第七屆學(xué)術(shù)討論會論文集（下冊）[C];2007年

2 王忠彥;胡林金;張明煊;劉景時;;喀喇昆侖山葉爾羌河50年來水資源演變特征[A];青藏高原資源·環(huán)境·生態(tài)建設(shè)學(xué)術(shù)研討會暨中國青藏高原研究會2007學(xué)術(shù)年會論文摘要匯編[C];2007年

3 廖福元;;基于經(jīng)驗?zāi)Ｊ椒纸獾姆瞧椒�(wěn)信號趨勢檢測方法[A];第六屆全國信息獲取與處理學(xué)術(shù)會議論文集（3）[C];2008年

4 王書平;李建平;高麗君;趙茜;;標(biāo)準(zhǔn)X-11方法及國際原油價格季節(jié)性波動分析[A];現(xiàn)代工業(yè)工程與管理研討會會議論文集[C];2006年

5 鄭祚芳;張秀麗;曹鴻興;謝莊;潘家華;;用去趨勢漲落分析研究北京氣候的長程變化特征[A];2008年北京氣象學(xué)會科技優(yōu)秀論文集[C];2008年

6 謝平;陳麗;唐亞松;胡彩霞;;變化環(huán)境下基于非線性趨勢分析的干旱評價方法[A];變化環(huán)境下的水資源響應(yīng)與可持續(xù)利用——中國水利學(xué)會水資源專業(yè)委員會2009學(xué)術(shù)年會論文集[C];2009年

7 鐘水映;李魁;;東亞地區(qū)工業(yè)化、城鎮(zhèn)化與農(nóng)地非農(nóng)化的協(xié)動性研究[A];節(jié)約集約用地及城鄉(xiāng)統(tǒng)籌發(fā)展——2009年海峽兩岸土地學(xué)術(shù)研討會論文集[C];2009年

相關(guān)重要報紙文章 前5條

1 國泰君安 姜超;將3年期左右債券作為核心配置[N];中國證券報;2008年

2 大摩投資 徐勝治;牛熊轉(zhuǎn)換 市場開打資源爭奪仗[N];證券日報;2006年

3 大摩投資 徐勝治;長期轉(zhuǎn)折發(fā)端 短期震蕩盤整[N];證券日報;2006年

4 東方趨勢 趙笑云;中級行情呼之欲出[N];福建工商時報;2001年

5 水皮;胳膊再粗?jǐn)Q不過大腿　[N];中華工商時報;2002年

相關(guān)博士學(xué)位論文 前6條

1 孫曉濤;趨勢周期分解理論與我國經(jīng)濟的周期[D];華中科技大學(xué);2013年

2 劉俊生;我國經(jīng)濟周期波動的測定和分析[D];吉林大學(xué);2007年

3 李慶華;我國經(jīng)濟周期階段性和波動性的動態(tài)計量研究[D];吉林大學(xué);2008年

4 王金明;我國轉(zhuǎn)軌時期經(jīng)濟周期波動的特征分析及監(jiān)測方法的應(yīng)用研究[D];吉林大學(xué);2006年

5 許斌;變化環(huán)境下區(qū)域水資源變異與評價方法不確定性[D];武漢大學(xué);2013年

6 楊淑霞;中國電力需求周期演變規(guī)律及轉(zhuǎn)折點研究[D];華北電力大學(xué)（河北）;2006年

相關(guān)碩士學(xué)位論文 前10條

1 唐敏;中國實際GDP的趨勢及周期成分分解[D];華中科技大學(xué);2011年

2 孟彩俠;基于不同方法的和田綠洲水循環(huán)要素變化特征研究[D];西安理工大學(xué);2006年

3 朱文金;數(shù)據(jù)預(yù)處理在預(yù)測模型中的應(yīng)用[D];蘭州大學(xué);2010年

4 馬亞男;股票收益率與通貨膨脹率的關(guān)聯(lián)性研究[D];吉林大學(xué);2008年

5 陳楠楠;城域網(wǎng)應(yīng)用層VoIP流量的建模與預(yù)測研究[D];湖南大學(xué);2009年

6 趙文勝;投機性短期國際資本流動對中國經(jīng)濟影響的實證研究[D];吉林大學(xué);2009年

7 夏冰瑩;證券市場資產(chǎn)價格的動力學(xué)關(guān)聯(lián)性研究[D];南京信息工程大學(xué);2012年

8 王娜娜;我國固定資產(chǎn)投資周期與經(jīng)濟周期波動的關(guān)聯(lián)性研究[D];吉林大學(xué);2007年

9 卿山鳳;企業(yè)價值評估中現(xiàn)金流量的模擬分析[D];西南交通大學(xué);2005年

10 翟子鈞;基于時間序列的我國鋼材以及有色金屬供給量分析[D];中國人民大學(xué);2008年

，

本文編號：2120417