1.Local Whittle Memory Parameter Estimation of Long Memory Process in the Presence of Low Frequency Contaminations (Job Market Paper) (Jie Hou and Pierre Perron) Revise and Resubmit at the Journal of Econometrics download
We propose a modified local-Whittle estimator of the memory parameter of a long memory time series process which has good properties under an almost complete collection of contamination processes that have been discussed in the literature, mostly separately. These contaminations include processes whose spectral density functions dominate at low frequencies such as random level shifts, deterministic level shifts and deterministic trends. We show that our modified estimator has the usual asymptotic distribution applicable for the standard local Whittle estimator in the absence of such contaminations. We also show how the estimator can be modified to further account for short memory dynamics and additive noise. Through extensive simulations, we show that the proposed estimator has substantial efficiency gains compared to existing semi parametric estimators and still remain valid for cases where the current estimators fail to apply. We analyze the effects of bandwidth selection on the performance of the estimators under different contaminations and the tradeoffs between bias and RMSE among those estimators. We recommend general rules to select the bandwidth and the exact form of the estimator in order to minimize the bias and RMSE.
2. Memory Parameter Estimation of Financial Time Series Robust to Low Frequency Contaminations (preliminary draft)
We use the semi parametric local Whittle estimator proposed in the paper “Local Whittle Memory Parameter Estimation of Long Memory Process in the Presence of Low Frequency Contaminations” to estimate the long-memory parameter of the stochastic volatility in the S&P 500 returns. We find that both long memory and low frequency contaminations are present using both high and low frequency volatility series. Our results contrast with some findings recently reported. We argue that previous studies obtained unreliable results under low frequency contaminations because they ignored the presence of additive noise, thereby inducing a downward bias in the estimates.
Work in Progress
1. “A Pivotal Test Statistic against Structural Changes in Multi-Equation Regression Systems”(with Pierre Perron)
2.“Using Interactive Effects Factor Analysis to Improve the Berry, Levinsohn and Pakes (1995) model used in industrial organization”
Hou, Jie and Yang, Lijun (2006) “Triangularization of a class of C1 unipotent maps,” the Fifth International Conference on Information and Management Sciences, Chengdu, China (ISTP Indexed).