Download presentation
Presentation is loading. Please wait.
Published byNaomi Poor Modified over 10 years ago
1
1 Conditional Forecasts in DSGE models Author: Junior Maih Discussant: Alon Binyamini Central Bank Macroeconomic Modeling Workshop 2009, Jerusalem
2
2 Outline Review: Motivation Technique Main results Comments on: Technique Implementation Conclusion Text
3
3 Review
4
4 Review – motivation Background – improving DSGE forecast by conditioning information. DSGE offer interpretation. DSGE are forward looking – future info is relevant and useful. Yet, may fail to forecast some (central) endogenous variables. Questions: How to carry out (soft) conditional forecast? When hard is superior to soft condition?
5
5 Review – contributions The Junior Smoother Technique for soft conditional forecast with fw-looking DSGE Nests hard and unconditional. Deals with two source of uncertainty (present and future) state uncertainty. Structural uncertainty. Extract the most likely shocks-combination that satisfies the conditioning restrictions. E.g. : conditioning on future interest rate … The spirit of Kalman smoother. Extension of earlier contribution - Waggoner & Zha (1999). Utilized to show that relevant information can be too tight.
6
6 Comments
7
7 Comment on the methodology Comment: Backcasting doesn’t employ the conditioning information. Back-casting is inefficient. Starting values are inconsistent with conditioning information. Not an issue for unconditional forecast. Suggestion – iterate till convergence: Compute conditional forecast using the Junior smoother. Extend sample by forecast. Update back-casting by Kalman smoother. Now, starting values for forecast may differ. So, repeat to convergence.
8
8 Comment on the implementation Verdict: Conditioning doesn ’ t necessarily improve forecast: Relevant information can be too tight. Disappointing result attributed to misspecification. But, Unconditional forecast is also misspecified. Ex-post realizations vis-à-vis ex-ante expectations. Two suggestions Repeat the analysis with ex-ante expectations. Divide and conquer by Monte-Carlo simulation: Forecast with “true” DGP (misspecification or irrelevant conditioning?). Distinguish between state and structural uncertainties.
9
9 Comparison with Kalman Smoother If not equivalent, where the does the difference come from? Is it equivalent to KS with conditioning restrictions as observables and nowcast for initialization? “…allows for the possibility of agents reacting to anticipated future events beyond one step ahead.” But this should be attributed to the state-space representation: “…it does not change the initial conditions of the state vector.” Can restrict the KS similarly. But, is it efficient? Keynes: “When the facts change, I change my mind. What do you do, sir?”
10
10 Comments on the text (cont’) Under similar treatment, would KS & JS extract similar or different shocks? Why? Chapter 2 (the intuition-building example) Why tighter conditioning shifts expectations without increasing certainty?
11
11 To conclude Two birds: Interesting. Useful. Appealing the disappointing verdict: Relevant and reliable hard conditioning might turn out to be too tight. But, ex-post realizations may not be relevant. Even if correct, misspecification might be the wrong guy to blame.
12
12 The End ! Thank you
13
13 The Bivariate normal example
Similar presentations
© 2025 SlidePlayer.com. Inc.
All rights reserved.