دانلود رایگان مقاله لاتین پیش بینی دولت ایالات متحده از سایت الزویر
عنوان فارسی مقاله:
اظهار نظر در مورد '' چگونگی پیش بینی دولت ایالات متحده از بدهی دولت فدرال
عنوان انگلیسی مقاله:
Comment on ‘‘How Biased are US Government Forecasts of the Federal Debt?’’
سال انتشار : 2017
برای دانلود رایگان مقاله پیش بینی دولت ایالات متحده اینجا کلیک نمایید.
بخشی از مقاله انگلیسی:
3. The IIS methodology
level until this adjustment in significance levels is made. By adjusting our significance level to 0.05, we are able to include 2002, 2003, and 2010 in our potential dummy variables as well. When these are included, all six dummy variables are found to be significant at the 0.02 level when the dummy variables are applied with the entire dataset, with all but 2003 and 2010 being significant at the 0.005 level. Using this bare-bones approach, we are unable to detect 1990 as a significant dummy variable, but we attribute this to our choice of blocks. Thus, we are led to conclude that the choices of blocks and significance levels used are essential for determining the final model when using the IIS technique.We investigate the sensitivity of the IIS technique by applying it to the one-year-ahead forecast errors for the CBO debt forecasts using data from 1984 to 2012. We imitate the bare-bones approach that Ericsson provided as an illustrative example by dividing this 29 year period into two parts, the first of 14 years and the second of 15 years. Note that Autometrics applies the IIS technique with many possibly unequally sized blocks. Because this simpler barebones approach uses only one partition, the number of points detected as being significant should be less than or equal to the number detected by the Autometrics software, but will provide a baseline for comparison. Using this bare bones set-up, we use the IIS model to detect years where the observed value was significantly different from the other years using a significance level of 0.01. We find only 2001, 2008, and 2009 to be significant at the 0.01 level when dummy variables are applied to the separate parts of the dataset, and only 2008 remains significant at this level when these three years are used as dummy variables with the entire dataset. In the Autometrics software output that was provided to us, we note that the significance level used appears to increase when performing an extended block search for omitted variables. We also note that the dummy variables that Autometrics originally selected (2001, 2002, 2003, 2008, and 2009) are not all significant at the 0.01
برای دانلود رایگان مقاله پیش بینی دولت ایالات متحده اینجا کلیک نمایید.
کلمات کلیدی:
22222
