We show that the conditional risk estimation in the ICAPM model (Merton, 1973) should contain the unspanned uncertainty beyond stock market if the interest rate is not sufficient to describe the dynamic investment state. Borrowing an aggregated uncertainty measure that captures unspanned...

For the popular mean-variance portfolio choice problem in the case without a risk-free asset, we develop a new portfolio strategy to mitigate estimation risk. We show that in both calibrations and real datasets, optimally combining the sample global minimum variance portfolio with a sample...

In this paper, we propose a cross-sectional option momentum strategy that is based on the risk component of delta-hedged option returns. We find strong evidence of risk continuation in option returns. Specifically, options with a high risk component significantly outperform those with a low risk...

We establish the out-of-sample predictability of monthly exchange rate changes via machine learning techniques based on 70 predictors capturing country characteristics, global variables, and their interactions. To guard against overfitting, we use the elastic net to estimate a high-dimensional...

We investigate an important question for institutional investors — namely, which hedge fund investing styles help to hedge against bad times? We define good versus bad times as (1) up and down equity market regimes derived from the 200-day moving average of the S&P 500 price index or (2)...

Using time-series trends of a set of firms' major fundamentals, we find that there is a fundamentalmomentum in the stock market. Buying stocks in the top quintile of fundamental trends and selling stocks in the bottom quintile earns a monthly average return of 0.88%, whose magnitude is...

Time series momentum (TSM) refers to the predictability of the past 12-month return on the next one-month return and is the focus of several recent influential studies. This paper shows that asset-by-asset time series regressions reveal little evidence of TSM, both in- and out-of-sample. While...

In this paper, we study portfolio choice problem under estimation risk and show why the 1/N rule is very difficult to beat in applications and studies. First, as long as the dimensionality is high relative to sample size, we show that the usual estimated investment strategies are biased even...

While common machine learning algorithms focus on minimizing the mean-square errors of model fit, we show that genetic programming, GP, is well-suited to maximize an economic objective, the Sharpe ratio of the usual spread portfolio in the cross-section of expected stock returns. In contrast to...

Based on intraday data for a large cross section of individual stocks, we find that the risk component of stock returns exhibits strong intraday momentum, and this pattern holds from previous market close to 10:00, and every half hour since then until market close at 16:00. Strikingly, the...