دانلود رایگان مقاله الزویر در مورد بررسی رویکرد قدرتمند پیش بینی (کلیک کنید)
توضیحات :
وبسایت مکاله اقدام به ارائه پروژه ی ساینس دایرکت با فرمت pdf، از انتشارات الزویر برای رشته علوم اقتصادی، و با عنوان بررسی رویکرد قدرتمند پیش بینی ، نموده است.
مشخصات این مقاله :
عنوان مقاله :
Robust approaches to forecasting
ترجمه فارسی عنوان :
بررسی رویکرد قدرتمند پیش بینی
سال انتشار : 2014
متعلق به مجله یا ژورنال : نشریه بین المللی پیش بینی – International Journal of Forecasting
فرمت: PDF
تعداد صفحات: 14
شماره پروژه: 5095
کلمات کلیدی :
Forecast biases,Smoothed forecasting devices,Factor models,GDP forecasts,Location shifts
پیش بینی تعصبات، دستگاه پیش بینی هموار، مدل های عامل، پیش بینی تولید ناخالصی داخلی، تغییرات محل سکونت
چکیده :
Abstract
We investigate alternative robust approaches to forecasting, using a new class of robust devices, contrasted with equilibrium-correction models. Their forecasting properties are derived facing a range of likely empirical problems at the forecast origin, including measurement errors, impulses, omitted variables, unanticipated location shifts and incorrectly included variables that experience a shift. We derive the resulting forecast biases and error variances, and indicate when the methods are likely to perform well. The robust methods are applied to forecasting US GDP using autoregressive models, and also to autoregressive models with factors extracted from a large dataset of macroeconomic variables. We consider forecasting performance over the Great Recession, and over an earlier more quiescent period.
مقدمه این مقاله :
Introduction
In recent times, there has been increased interest in forecasting with diffusion indices and factor models: see e.g., Castle, Clements, and Hendry (2013), Forni, Hallin,Lippi, and Reichlin (2000), Peña and Poncela (2004), Schumacher and Breitung (2008) and Stock and Watson (1989,1999, 2009).3 In Castle, Clements, and Hendry (2013), we investigated which approach to forecasting output levels and growth using factors, variables, both, or neither performed best on quarterly data over the Great Recession to 2011. After updating and extending the dataset from Stock and Watson (2009), we used Autometrics (as described in Doornik (2009), and Doornik and Hendry (2013))for in-sample modeling, which allowed all the principal components of the variables as well as the original variables to be included jointly, while also tackling multiple breaks by impulse-indicator saturation (IIS: see Castle, Doornik, & Hendry, 2012, and Johansen & Nielsen,2009). Forecasting US GDP growth over 1-, 4- and 8-step horizons showed that factor models were somewhat more useful for short-term forecasting, but their relative performance declined as the forecast horizon increased. We found (like many other investigators) that it was difficult to beat scalar autoregressions: Fildes and Stekler (2002) provide a survey of macroeconomic forecasting before the Great Recession, which the follow up in Stekler and Talwar (2011) show was not well predicted. Our own forecasts for GDP levels highlighted the need for robust strategies (such as intercept corrections) when location shifts (i.e., shifts in the previous unconditional mean) occurred.
توجه :
– این مقاله به صورت کامل و با فرمت پی دی اف آماده خرید اینترنتی و دانلود آنی میباشد.
توضیحات بیشتر در مورد پروژه :
در این مقاله خواهید خواند که روش جدیدی با استفاده از یک کلاس و دستگاه های قوی جایگزین مدل قبلی شده است. از نظر ما پیش بینی عملکرد پیش از رکود بزرگ رایج تر می باشد.
دانلود رایگان مقاله الزویر در مورد بررسی رویکرد قدرتمند پیش بینی (کلیک کنید)
