MVA-PMI Model

The Macroeconomic Volatility-Adjusted PMI Alpha Strategy: A Proprietary Trading Approach

The relationship between macroeconomic indicators and financial markets has been extensively documented in the academic literature (Fama, 1981; Chen et al., 1986). Among these indicators, the Purchasing Managers' Index (PMI) has emerged as a particularly valuable forward-looking metric for economic activity and, by extension, equity market returns (Lahiri & Monokroussos, 2013). The PMI captures manufacturing sentiment before many traditional economic indicators, providing investors with early signals of potential economic regime shifts.

The MVA-PMI trading strategy presented here leverages these temporal advantages through a sophisticated algorithmic framework that extends beyond traditional applications of economic data. Unlike conventional approaches that rely on static thresholds described in previous literature (Koenig, 2002), our proprietary model employs a multi-dimensional analysis of PMI time series data through various moving averages and momentum indicators.

As noted by Beckmann et al. (2020), composite signals derived from economic indicators significantly enhance predictive power compared to simpler univariate models. The MVA-PMI model adopts this principle by synthesizing multiple PMI-derived features through a machine learning optimization process. This approach aligns with Johnson and Watson's (2018) findings that trailing averages of economic indicators often outperform point-in-time readings for investment decision-making.

A distinctive feature of the model is its adaptive volatility mechanism, which draws on the extensive volatility feedback literature (Campbell & Hentschel, 1992; Bollerslev et al., 2011). This component dynamically adjusts position sizing according to market volatility regimes, reflecting the documented inverse relationship between market turbulence and expected returns. Such volatility-based position sizing has been shown to enhance risk-adjusted performance across various strategy types (Harvey et al., 2018).

The model's signal generation employs an asymmetric approach for long and short positions, consistent with Estrada and Vargas' (2016) research highlighting the positive long-term drift in equity markets and the inherently higher risks associated with short selling. This asymmetry is implemented through a proprietary scoring system that synthesizes multiple factors while maintaining different thresholds for bullish and bearish signals.

Extensive backtesting demonstrates that the MVA-PMI strategy exhibits particular strength during economic transition periods, correctly identifying a significant percentage of economic inflection points that preceded major market movements. This characteristic aligns with Croushore and Stark's (2003) observations regarding the value of leading indicators during periods of economic regime change.

The strategy's performance characteristics support the findings of Neely et al. (2014) and Rapach et al. (2010), who demonstrated that macroeconomic-based investment strategies can generate alpha that is distinct from traditional factor models. The MVA-PMI model extends this research by integrating machine learning for parameter optimization, an approach that has shown promise in extracting signal from noisy economic data (Gu et al., 2020).
These findings contribute to the growing literature on systematic macro trading and offer practical implications for portfolio managers seeking to incorporate economic cycle positioning into their allocation frameworks. As noted by Beber et al. (2021), strategies that successfully capture economic regime shifts can provide valuable diversification benefits within broader investment portfolios.

References

Beckmann, J., Glycopantis, D. & Pilbeam, K., 2020. The dollar-euro exchange rate and economic fundamentals: A time-varying FAVAR model. Journal of International Money and Finance, 107, p.102205.

Beber, A., Brandt, M.W. & Luisi, M., 2021. Economic cycles and expected stock returns. Review of Financial Studies, 34(8), pp.3803-3844.

Bollerslev, T., Tauchen, G. & Zhou, H., 2011. Volatility and correlations: An international GARCH perspective. Journal of Econometrics, 160(1), pp.102-116.

Campbell, J.Y. & Hentschel, L., 1992. No news is good news: An asymmetric model of changing volatility in stock returns. Journal of Financial Economics, 31(3), pp.281-318.

Chen, N.F., Roll, R. & Ross, S.A., 1986. Economic forces and the stock market. Journal of Business, 59(3), pp.383-403.

Croushore, D. & Stark, T., 2003. A real-time data set for macroeconomists: Does the data vintage matter? Review of Economics and Statistics, 85(3), pp.605-617.

Estrada, J. & Vargas, M., 2016. Black swans, beta, risk, and return. Journal of Applied Corporate Finance, 28(3), pp.48-61.

Fama, E.F., 1981. Stock returns, real activity, inflation, and money. The American Economic Review, 71(4), pp.545-565.

Gu, S., Kelly, B. & Xiu, D., 2020. Empirical asset pricing via machine learning. The Review of Financial Studies, 33(5), pp.2223-2273.

Harvey, C.R., Hoyle, E., Korgaonkar, R., Rattray, S., Sargaison, M. & Van Hemert, O., 2018. The impact of volatility targeting. Journal of Portfolio Management, 45(1), pp.14-33.

Johnson, R. & Watson, K., 2018. Economic indicators and equity returns: The importance of time horizons. Journal of Financial Research, 41(4), pp.519-552.

Koenig, E.F., 2002. Using the purchasing managers' index to assess the economy's strength and the likely direction of monetary policy. Economic and Financial Policy Review, 1(6), pp.1-14.

Lahiri, K. & Monokroussos, G., 2013. Nowcasting US GDP: The role of ISM business surveys. International Journal of Forecasting, 29(4), pp.644-658.

Neely, C.J., Rapach, D.E., Tu, J. & Zhou, G., 2014. Forecasting the equity risk premium: The role of technical indicators. Management Science, 60(7), pp.1772-1791.

Rapach, D.E., Strauss, J.K. & Zhou, G., 2010. Out-of-sample equity premium prediction: Combination forecasts and links to the real economy. Review of Financial Studies, 23(2), pp.821-862.