首页--经济论文--经济计划与管理论文--经济计算、经济数学方法论文--经济统计学论文

Alternative Statistical Methods to Forecast Chinas Gnp

英文摘要第7页
CONTENTS第9-14页
LIST OF FIGURES第14-16页
LIST OF TABLES第16-17页
PREFACE第17-18页
GLOSSARY第18-20页
INTRODUCTION第20-30页
    1- The aim of this thesis第21-22页
    2- Hypotheses第22页
    3- the Importance of this thesis第22-23页
    4- The limitations of this thesis第23-24页
    5- The previous studise第24-29页
    6- The data at this thesis第29页
    7- The statistical computer package第29-30页
THE THEORETICAL STUDY第30-79页
    1- Introduction第31页
    2- The regression analysis第31-50页
        2-1 Model assumptions第31-32页
        2-2 The least squares estimates(OLS)第32页
        2-3 Hypothesis testing in regression analysis第32-34页
        2-4 Some regression's problems第34-39页
            2-4-1 The multicollinearity第34-35页
            2-4-2 Autocorrelation第35-39页
        2-5 Stepwise model selection第39-40页
        2-6 Principal components第40-45页
        2-7 Lagged variables第45-49页
        2-8 Chow's second test and predictive failure第49-50页
    3- The univariate time series analysis第50-58页
        3-1 Stationarity第51-54页
            3-1-1 Testing for stationarity第51-52页
            3-1-2 Transforming to stationary第52-54页
                3-1-2-1 Stabilizing variances第52-54页
                3-1-2-2 Stabilizing the mean第54页
        3-2 The Box-Jenkins approach第54-58页
            3-2-1 Identification第55-56页
            3-2-2 Estimation第56页
            3-2-3 Diagnostic checking第56-58页
                3-2-3-1 Akaike Information Criterion(AIC)第57页
                3-2-3-2 Schwarz Bayesian Criterion(SBC)第57-58页
                3-2-3-3 Log Likelihood Criterion(Log LH)第58页
    4- The multivariate time series analysis第58-64页
        4-1 The transfer function model第58-61页
        4-2 Stationarity第61页
        4-3 Identification of the transfer function models第61-62页
        4-4 Estimation第62-63页
        4-5 Diagnostic checking第63-64页
    5- Combine regression analysis and time series analysis第64-74页
        5-1 The method of building an ARIMA model for the error series第64-65页
            5-1-1 Estimating the regression model第64-65页
            5-1-2 Building an ARIMA model for the residual第65页
        5-2 Re-estimation the regression model when autocorrelated disturbances are found第65-74页
            5-2-1 Maximum likelihood estimation第70-73页
            5-2-2 Prediction in the presence of autocorrelated disturbances第73-74页
    6- Comparison study to the entire suggested model in this thesis第74-79页
        6-1 Some statistical criteria for comparison第74-75页
            6-1-1 The criterion of the mean absolute relative prediction error(MARPE):第74页
            6-1-2 The criterion of the mean squared relative prediction error(MSRPE)第74-75页
            6-1-3 The criterion of the root mean squared prediction error(RMSPE)第75页
        6-2 Contingency chi-square goodness-of-fit test第75-76页
        6-3 Two-sample sign test第76-77页
        6-4 Rank-sum test(Wilcoxon test)第77-79页
THE PRACTICAL STUDY第79-155页
    1- Applying the regression analysis using explanatory variables第81-92页
        1-1 Estimate the model第81-85页
        1-2 Re-estimate the regression models using stepwise method第85-87页
        1-3 Comparing the models第87-90页
            1-3-1 Check the model's predicted values第87-88页
            1-3-2 Comparison the models in the second group第88页
            1-3-3 Comparison the models in the first and second group第88-90页
        1-4 Results of evaluating the logarithmic model's ability to forecast第90-91页
            1-4-1 Test of parameter stability(Chow's second test)第90-91页
        1-5 Conclusion第91-92页
    2- Applying the Regression Analysis Using the Principal Component第92-100页
        2-1 Factor analysis result第92页
        2-2 Results of the regression analysis第92-94页
        2-3 Comparing the models第94-96页
            2-3-1 Check the model's predicted values第95页
            2-3-2 Comparison of the models in the first and second group第95-96页
        2-4 Results of evaluating the linear model's ability to forecast第96-99页
            2-4-1 Test of parameter stability(Chow's second test)第96-99页
        2-5 Conclusion第99-100页
    3- Applying the Regression Analysis Using the Lag Variables with other Explanatory Variables第100-106页
        3-1 Estimating the model第100页
        3-2 Testing for autocorrelation第100-103页
            3-2-1 Runs test第101-102页
            3-2-2 Durbin's h test第102-103页
        3-3 Results of evaluating the lag model's ability to forecast第103-105页
        3-4 Conclusion第105-106页
    4- Applying the Univariate Time Series Analysis第106-117页
        4-1 Testing and Transforming to Stationarity第106-111页
            4-1-1 Testing for Stationarity第106-107页
            4-1-2 Transforming to stationary第107-111页
                4-1-2-1 Stabilizing variances第107-110页
                4-1-2-2 Stabilizing the mean第110-111页
        4-2 Identification第111-114页
        4-3 Estimating the model第114页
        4-4 Model's diagnostic checking第114-116页
        4-5 Conclusion第116-117页
    5- Applying the multivariate time series analysis第117-137页
        5-1 Testing and transforming Total Exports data to stationary第117-121页
            5-1-1 Testing for stationarity第117-118页
            5-1-2 Transforming to Stationary第118-121页
                5-1-2-1 Stabilizing variances第119-120页
                5-1-2-2 Stabilizing the mean第120-121页
        5-2 Testing and Transforming Total Imports Data to Stationary第121-126页
            5-2-1 Testing for Stationarity第121-123页
            5-2-2 Transforming to Stationary第123-126页
                5-2-2-1 Stabilizing variances第123-125页
                5-2-2-2 Stabilizing the mean第125-126页
        5-3 Identification第126-129页
        5-4 Estimate the Multivariate Time Series Model第129-132页
        5-5 Diagnostic checking of multivariate time series models第132-136页
        5-6 Conclusion第136-137页
    6- Combine the regression analysis with the time series analysis第137-145页
        6-1 Building the ARIMA model for the residual series第137-145页
            6-1-1 Testing and Transforming to Stationarity第137-138页
                6-1-1-1 Testing for Stationarity第137-138页
            6-1-2 Identification第138-140页
            6-1-3 Estimation第140页
            6-1-4 The Diagnostic Checking第140-145页
    7- Another way to combine time series and regression analysis第145-150页
        7-1 Maximum likelihood estimation第145-146页
        7-2 Cochrane-Orcutt estimation第146-147页
        7-3 Prais-Winsten estimation第147页
        7-4 Diagnostic checking第147-149页
        7-5 Conclusion第149-150页
    8- Comparison study to all the suggested model in this thesis第150-155页
        8-1 Re-introduce the best seven models第150-151页
        8-2 Comparing the regression models第151-152页
        8-3 Plotting the valid models with the actual values of GNP第152-153页
        8-4 Summarizing the diagnostic checking results第153-155页
CONCLUSIONS AND RECOMMENDATIONS第155-159页
    1- The Conclusions第156-158页
    2- The recommendations第158-159页
REFERENCES第159-160页

论文共160页,点击 下载论文
上一篇:时间导向与主观幸福感对享乐主义/功利主义的影响研究
下一篇:AJ信托公司结构化证券信托业务及风险管理研究