TY - JOUR
T1 - Risks for the long run
T2 - Estimation with time aggregation
AU - Bansal, Ravi
AU - Kiku, Dana
AU - Yaron, Amir
N1 - Funding Information:
We thank Andy Abel, George Constantinides, Lars Hansen, John Heaton, Ricardo Reiss (editor) Tom Sargent, Martin Schneider, Annette Vissing-Jorgensen, anonymous referee, seminar participants at Boston University, Copenhagen School of Business, Carnegie-Mellon University, Duke University, Harvard University, LBS, LSE, Norwegian School of Management, NYU, Stockholm School of Economics, Tel-Aviv University, University of California-Berkeley, University of Chicago, University of Texas-Austin, University of Washington-St. Louis, University of Wisconsin, Wharton, Yale, and conference participants at the Summer Econometric Society Meetings, the NBER Asset-Pricing meeting, Nemmers Prize Conference, Frontiers of Monetary policy – St. Louis FRB, and CIRANO Financial Econometrics Conference in Montreal for useful comments. Yaron thanks support from the Rodney White Center at the Wharton School. This paper builds on our earlier working paper, Risks for the Long Run: Estimation and Inference.
Publisher Copyright:
© 2016
PY - 2016/9/1
Y1 - 2016/9/1
N2 - The discrepancy between the decision and data-sampling intervals, known as time aggregation, confounds the identification of long-, short-run growth, and volatility risks in asset prices. This paper develops a method to simultaneously estimate the model parameters and the decision interval of the agent by exploiting identifying restrictions of the Long Run Risk (LRR) model that account for time aggregation. The LRR model finds considerable empirical support in the data; the estimated decision interval of the agents is 33 days. Our estimation results establish that long-run growth and volatility risks are important for asset prices.
AB - The discrepancy between the decision and data-sampling intervals, known as time aggregation, confounds the identification of long-, short-run growth, and volatility risks in asset prices. This paper develops a method to simultaneously estimate the model parameters and the decision interval of the agent by exploiting identifying restrictions of the Long Run Risk (LRR) model that account for time aggregation. The LRR model finds considerable empirical support in the data; the estimated decision interval of the agents is 33 days. Our estimation results establish that long-run growth and volatility risks are important for asset prices.
KW - Asset prices
KW - Long-run risks
KW - Time-aggregation
UR - http://www.scopus.com/inward/record.url?scp=84979656078&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=84979656078&partnerID=8YFLogxK
U2 - 10.1016/j.jmoneco.2016.07.003
DO - 10.1016/j.jmoneco.2016.07.003
M3 - Article
AN - SCOPUS:84979656078
SN - 0304-3932
VL - 82
SP - 52
EP - 69
JO - Carnegie-Rochester Confer. Series on Public Policy
JF - Carnegie-Rochester Confer. Series on Public Policy
ER -