Return Predictability and Market-Timing: A One-Month Model by Petra Bakosova

Petra Bakosova, Chief Operating Officer of Hull Tactical, presented her paper, “Return Predictability and Market-Timing: A One-Month Model,” at QuantCon 2018. Petra covered the history of equity risk premium modeling and market timing, and then summarized the findings of the research paper. You can find her video from QuantCon 2018 below, supplemented by her summary.

Return Predictability and Market-Timing: A One-Month Model
The presentation revolved around a one-month market-timing model constructed from 15 diverse variables. Authors used weighted least squares with stepwise variable selection to build a predictive model for the one-month-ahead market excess returns, and transformed the model’s forecasts into investable positions to build a market-timing strategy. From 2003 to 2017, this strategy results in 16.6% annual returns with a 0.92 Sharpe ratio and a 20.3% maximum drawdown, whereas the S&P 500 has annual returns of 10%, a 0.46 Sharpe ratio, and a maximum drawdown of 55.2%. Lastly, the combination of the one-month model with the market timing models of Hull and Qiao (2017) and Hull, Bakosova and Kment (2018) resulted in a Sharpe ratio that exceeds the individual model Sharpe ratios. Updated forecasts from our one-month model are updated in Hull Tactical’s Daily Report.

Variables
Petra discussed the background and intuition behind the following candidate variables:

• Change in Inflation
• Industrial Production
• Slope of the Interest Rate Term Structure
• Commodity Prices
• Housing Starts
• Exchange Rate
• FRB Loan Officer Survey
• Delinquencies
• New Orders and New Shipments
• Baltic Dry Index
• National Association of Purchasing Managers
• Change in Unemployment Rate
• Momentum
• PCA of Price Ratios

Model
The goal of this paper was to build a predictive model for the one-month future market excess returns and construct a market-timing strategy. Some of the 15 considered predictors have been proposed in the predictability literature, whereas others were unique to our paper. Weighted least squares (WLS) regression with stepwise variable selection is used to forecast next month’s market excess returns. The stepwise WLS puts zero weight on marginal variables that do not add substantially to the model. Within each month, model parameters are held constant and forecasts are formed using updated predictor values each day. The authors then construct a monthly market-timing strategy by transforming forecast values into investable positions. The restrictions placed on the transformation include 1) 0% invested in the S&P 500 and 100% invested in T-bills when the equity premium forecast is zero, 2) 100% invested in the S&P 500 when the forecast of the equity premium is equal to its historical average, and 3) maximum of 150% exposure to the S&P 500 when the equity premium forecast is above its historical average. After considering different ways we could make such a transformation, the authors chose a simple method that takes into account our confidence in the model. They scale the equity premium forecasts by the inverse of the Root Mean Square Error (RMSE) and multiply it by five to satisfy the above conditions. In particular, when the equity premium forecast is at its historical mean, scaling by the RMSE and multiplying by five results in a position close to 100%. When the equity premium forecast is zero, the model is 0% invested. Finally, the position is restricted to be between 0% and 150% invested in the S&P 500.

Takeaways
There is a preponderance of positive evidence for return predictability. The presented paper asserts that we can use our knowledge of return predictability to consistently produce returns which exceed those of buy-and-hold. The one-month market-timing model doubles the Sharpe ratio of the buy-and-hold strategy from 2003 to 2017 and greatly reduces the maximum drawdown. A combination of the one-month model, six-month model and the seasonal and trend model is more efficient than either of the models separately. A portfolio investing equal dollar amounts in the three market-timing strategies results in a higher Sharpe ratio and smaller drawdowns than both the one-month and the six-month strategies.

Return predictability has become well-accepted in the academic literature and has shown up in top business school finance courses. The continued stigma associated with market-timing primarily comes from the debate on whether predictability is sufficiently strong for a trading strategy to be built. Such a trading strategy would need to create enough economic value to offset its costs and produce superior returns compared to buy-and-hold. As more strategies such as the one we presented here develop, the attitude towards market-timing may change. Just as it was considered irresponsible to participate in market-timing in the last 30 years, it may be considered irresponsible not to participate in market-timing in the next 30 years.

©2018 Hull Tactical Asset Allocation, LLC (“HTAA”) is a Registered Investment Adviser.
The information set forth in HTAA's market commentaries and writings are of a general nature and are provided solely for the use of HTAA, its clients and prospective clients. This information is not intended to be and does not constitute investment advice. These materials reflect the opinion of HTAA on the date of production and are subject to change at any time without notice. Due to various factors, including changing market conditions or tax laws, the content may no longer be reflective of current opinions or positions. Past performance does not guarantee future results. All investments are subject to risks. Where data or information is presented that was prepared by third parties, such information will be cited and any such third-party sources have been deemed to be reliable. However, HTAA does not warrant or independently verify the accuracy of such information. HTAA and any third parties listed or identified herein, including The Journal of Portfolio Management, are separate and unaffiliated, are not responsible for each other’s products, policies or services, and the views expressed are their own.

Disclaimer

The material on this website is provided for informational purposes only and does not constitute an offer to sell, a solicitation to buy, or a recommendation or endorsement for any security or strategy, nor does it constitute an offer to provide investment advisory services by Quantopian. In addition, the material offers no opinion with respect to the suitability of any security or specific investment. No information contained herein should be regarded as a suggestion to engage in or refrain from any investment-related course of action as none of Quantopian nor any of its affiliates is undertaking to provide investment advice, act as an adviser to any plan or entity subject to the Employee Retirement Income Security Act of 1974, as amended, individual retirement account or individual retirement annuity, or give advice in a fiduciary capacity with respect to the materials presented herein. If you are an individual retirement or other investor, contact your financial advisor or other fiduciary unrelated to Quantopian about whether any given investment idea, strategy, product or service described herein may be appropriate for your circumstances. All investments involve risk, including loss of principal. Quantopian makes no guarantees as to the accuracy or completeness of the views expressed in the website. The views are subject to change, and may have become unreliable for various reasons, including changes in market conditions or economic circumstances.

3 responses

This approach seems to be at odds, to put it mildly, with the Quantopian market + everything else neutral strategy!

Regarding the third point under Model:
3) maximum of 150% exposure to the S&P 500 when the equity premium forecast is above its historical average.

Does 150% exposure mean 1.5 leverage? I don't see any other way to interpret what 150% exposure would mean otherwise.

@Evan, yes that is correct.