ForecastingFedFundsRate联邦基金利率预测

上传人:ra****d 文档编号:252557694 上传时间:2024-11-17 格式:PPT 页数:19 大小:298.50KB
返回 下载 相关 举报
ForecastingFedFundsRate联邦基金利率预测_第1页
第1页 / 共19页
ForecastingFedFundsRate联邦基金利率预测_第2页
第2页 / 共19页
ForecastingFedFundsRate联邦基金利率预测_第3页
第3页 / 共19页
点击查看更多>>
资源描述
Click to edit Master title style,*,Click to edit Master text styles,Second level,Third level,Fourth level,Fifth level,Forecasting Fed Funds Rate,Group 4,Neelima Akkannapragada,Chayaporn Lertrattanapaiboon,Anthony Mak,Joseph Singh,Corinna Traumueller,Hyo Joon You,Background,Fed funds rate(FFR)as an instrument of control.,FFR as sign of economic strength/weakness.,FFR is at 1.25%,the lowest since 1961.,Greenspan intimates at possibility of deflation(last week).,Japanese Deflation and the Great Depression.,Objectives,What will happen to the FFR given indicators such as GDP,CPI,stock market price levels,etc?,Create a distributed lag model with FFR as the dependent variable.,Provide one period ahead forecast of FFR.,And what does this forecast mean to us?,Provide economic context for the forecast.,The Idea,Supposing that the Fed made its decision solely on previous FFR would be naive.,Feds decision on future FFR depends on existing information.,We focus on these existing information to explain FFR.,GDP,CPI,SP500,Data Standardization,All data from Fred II.,Different time range and frequencies,But same time range and frequencies necessary for DL model,Lower bound set by data with the latest start(SP5000 Jan 1970),Upper bound set by data with the earliest end(GDP Jan 2003),Frequency set by data with lowest frequency(GDP quarterly).,Result is a shorter and less frequent data set(120 obs).,Still enough data.,Trace of Variables,Trace of Stationary Variables,Pairwise Granger Causality Tests,Date:05/27/03 Time:14:23,Sample:1970:1 2003:2,Lags:2,Null Hypothesis:ObsF-StatisticProbability,DLGDP does not Granger Cause DLFFR130 12.8145 8.7E-06,DLFFR does not Granger Cause DLGDP 1.75070 0.17788,DLSP does not Granger Cause DLFFR130 7.35499 0.00096,DLFFR does not Granger Cause DLSP 2.07473 0.12989,DDLCPI does not Granger Cause DLFFR129 0.61862 0.54034,DLFFR does not Granger Cause DDLCPI,7.36316 0.00095,DLSP does not Granger Cause DLGDP,130 1.16482 0.31534,DLGDP does not Granger Cause DLSP,0.54295 0.58240,DDLCPI does not Granger Cause DLGDP129 3.40096 0.03648,DLGDP does not Granger Cause DDLCPI 2.80740 0.06420,DDLCPI does not Granger Cause DLSP129 1.48890 0.22963,DLSP does not Granger Cause DDLCPI 0.48034 0.61972,Time Causality,Cross Correlogram 1,Cross Correlogram 2,Dependent Variable:DLFFR,Method:Least Squares,Sample(adjusted):1972:2 2003:1,Included observations:124 after adjusting endpoints,Convergence achieved after 8 iterations,Variable Coefficient Std.Error t-Statistic Prob.,4.6565699798e-07,DLGDP(-1)5.742530598661.3480009163 4.26003464036,AR(5)0.2469531144840.0878659566173 2.8105665037,0.00578442215627,R-squared0.27792384769 Mean dependent var -0.00836815797483,Sum squared resid2.07114473911 Schwarz criterion -1.05993683855,Log likelihood 77.766787904,F-statistic 11.4506405485,Durbin-Watson stat1.79038035224,Prob(F-statistic)6.75508422109e-08,Inverted AR Roots .76 .23+.72i .23-.72i -.61-.44i,-.61+.44i,Estimation Output DL Model,Residual Correlogram of the DL Model,Residual Diagnostics,Forecast,Summary,Standardization of data for DL modeling causes results in fewer observations.,Granger test is useful in isolating independent variables.,dlSP500 did not have AR structure.Creating the transformed dependent variable may have been more difficult.,Result is more plausible than ARMA model.,Fed funds rate will go down next quarter.,What Now?,Assuming that fed funds will continue to go down,one can,buy treasury bonds now and sell them later at a higher price when interest rate drops,simply try harder to find a job in the sluggish economy,start a business now in anticipation of next boom,Dependent Variable:DLFFR,Method:Least Squares,Sample(adjusted):1970:3 2003:1,Included observations:131 after adjusting endpoints,Convergence achieved after 3 iterations,VariableCoefficient Std.Error t-Statistic Prob.,C-0.014695 0.016465 -0.892533 0.3738,AR(1)0.166843 0.088232 1.890959 0.0609,R-squared 0.026971 Mean dependent var-0.014326,Adjusted R-squared 0.019428 S.D.dependent var 0.158538,S.E.of regression 0.156991 Akaike info criterion0.850112,Sum squared resid 3.179341 Schwarz criterion-0.806216,Log likelihood 57.68233 F-statistic 3.575725,Durbin-Watson stat 1.961340 Prob(F-statistic)0.060873,Estimation Output AR Model,Residual Correlogram AR Model,Residual of the AR Model,
展开阅读全文
相关资源
正为您匹配相似的精品文档
相关搜索

最新文档


当前位置:首页 > 商业管理 > 商业计划


copyright@ 2023-2025  zhuangpeitu.com 装配图网版权所有   联系电话:18123376007

备案号:ICP2024067431-1 川公网安备51140202000466号


本站为文档C2C交易模式,即用户上传的文档直接被用户下载,本站只是中间服务平台,本站所有文档下载所得的收益归上传人(含作者)所有。装配图网仅提供信息存储空间,仅对用户上传内容的表现方式做保护处理,对上载内容本身不做任何修改或编辑。若文档所含内容侵犯了您的版权或隐私,请立即通知装配图网,我们立即给予删除!