82 research outputs found
Prune Once for All: Sparse Pre-Trained Language Models
No full text@article{zafrir2021prune,
title={Prune Once for All: Sparse Pre-Trained Language Models},
author={Zafrir, Ofir and Larey, Ariel and Boudoukh, Guy and Shen, Haihao and Wasserblat, Moshe},
journal={arXiv preprint arXiv:2111.05754},
year={2021}
The Myth of Long-Horizon Predictability
No full textThe prevailing view in finance is that the evidence for long-horizon stock return predictability is significantly stronger than that for short horizons. We show that for persistent regressors, a characteristic of most of the predictive variables used in the literature, the estimators are almost perfectly correlated across horizons under the null hypothesis of no predictability. For example, for the persistence levels of dividend yields, the analytical correlation is 99% between the 1- and 2-year horizon estimators and 94% between the 1- and 5-year horizons, due to the combined effects of overlapping returns and the persistence of the predictive variable. Common sampling error across equations leads to ordinary least squares coefficient estimates and R2s that are roughly proportional to the horizon under the null hypothesis. This is the precise pattern found in the data. The asymptotic theory is corroborated, and the analysis extended by extensive simulation evidence. We perform joint tests across horizons for a variety of explanatory variables, and provide an alternative view of the existing evidence.
An Equilibrium Model of Nominal Bond Prices with Inflation-Output Correlation and Stochastic Volatility.
No full textA vector autoregressive (VAR) model is used to describe the joint dynamics of consumption growth and inflation. The commonly used homoscedastic VAR is extended to allow for stochastic volatility, driven by an unobservable autoregressive factor. Bond prices, the conditional expectation of a function of these factors, are approximated using Tauchen's quadrature method. We show that the mean, variance, and autocorrelation of yields is captured relatively well by the VAR-SV model, calibrated with inflation and consumption data. The co-dependents of consumption and inflation are shown to be important determinants for both real and nominal rates. Time variations in inflation volatility generate realistic variability of risk premia, but unrealistically low average magnitudes. Copyright 1993 by Ohio State University Press.
The Myth of Long-Horizon Predictability
No full textThe prevailing view in finance is that the evidence for long-horizon stock return predictability is significantly stronger than that for short horizons. We show that for persistent regressors, a characteristic of most of the predictive variables used in the literature, the estimators are almost perfectly correlated across horizons under the null hypothesis of no predictability. For the persistence levels of dividend yields, the analytical correlation is 99% between the 1- and 2-year horizon estimators and 94% between the 1- and 5-year horizons. Common sampling error across equations leads to ordinary least squares coefficient estimates and R-super-2s that are roughly proportional to the horizon under the null hypothesis. This is the precise pattern found in the data. We perform joint tests across horizons for a variety of explanatory variables and provide an alternative view of the existing evidence. The Author 2006. Published by Oxford University Press on behalf of The Society for Financial Studies. All rights reserved. For Permissions, please email: [email protected], Oxford University Press.
FINANCE RESEARCH SEMINAR SUPPORTED BY UNIGESTION "WHICH NEWS MOVES STOCK PRICES? A TEXTUAL ANALYSIS" WHICH NEWS MOVES STOCK PRICES? A TEXTUAL ANALYSIS 1
No full textAbstract A basic tenet of financial economics is that asset prices change in response to unexpected fundamental information. Since A basic tenet of financial economics is that asset prices change in response to unexpected fundamental information. Since Roll's (1988) provocative presidential address that showed little relation between stock prices and news, however, the finance literature has had limited success reversing this finding. This paper revisits this topic in a novel way. Using advancements in the area of textual analysis, we are better able to identify relevant news, both by type and by tone. Once news is correctly identified in this manner, there is considerably more evidence of a strong relationship between stock price changes and information. For example, market model R 2 s are no longer the same on news versus no news days (i.e., Roll's (1988) infamous result), but now are 16% versus 33%; variance ratios of returns on identified news versus no news days are 120% higher versus only 20% for unidentified news versus no news; and, conditional on extreme moves, stock price reversals occur on no news days, while identified news days show an opposite effect, namely a strong degree of continuation. A number of these results are strengthened further when the tone of the news is taken into account by measuring the positive/negative sentiment of the news story. 1 Corresponding author: Shimon Kogan, GSB 5.159, McCombs School of Business, University of Texas at Austin, 1 University Station, B6600, Austin, TX 78712, Tel: +1 (512) 232-6839, email [email protected]. We would like to thank John Griffin and seminar participants at the University of Texas, Austin, and Stern NYU for their comments and suggestions
A Multifactor, Nonlinear, Continuous-Time Model of Interest Rate Volatility
No full textThis paper presents a general, nonlinear version of existing multifactor models, such as Longstaff and Schwartz (1992). The novel aspect of our approach is that rather than choosing the model parameterization out of thin air,' our processes are generated from the data using approximation methods for multifactor continuous-time Markov processes. In applying this technique to the short- and long-end of the term structure for a general two-factor diffusion process for interest rates, a major finding is that the volatility of interest rates is increasing in the level of interest rates only for sharply upward sloping term structures. In fact, the slope of the term structure plays a larger role in determining the magnitude of the diffusion coefficient. As an application, we analyze the model's implications for the term structure of term premiums.
THE INFORMATION IN LONG-MATURITY FORWARD RATES: IMPLICATIONS FOR EXCHANGE RATES AND THE FORWARD PREMIUM ANOMALY
No full textThe forward premium anomaly is one of the most robust puzzles in financial economics.
We recast the underlying parity relation in terms of cross-country differences between
forward interest rates rather than spot interest rates with dramatic results. These forward
interest rate differentials have statistically and economically significant forecast power for annual exchange rate movements, both in- and out-of-sample, and the signs and magnitudes of the corresponding coefficients are consistent with economic theory. Forward interest rates also forecast future spot interest rates and future inflation. Thus,we attribute much of the forward premium anomaly to the anomalous behavior of short term interest rates, not to a breakdown of the link between fundamentals and exchange rates
Is the ex ante risk premium always positive? A new approach to testing conditional asset pricing models
No full textThis paper develops tests of inequality restrictions implied by conditional asset pricing models. The methodology is easy to implement, requires little knowledge of the conditional distribution of asset returns, and is valid under fairly weak assumptions. As an application, we test whether the ex ante risk premium is always positive. We report reliable evidence that the ex ante risk premium is negative in some states of the world; these states are related to periods of high expected inflation and especially to downward-sloping term structures
- …
