Our Services

Jantz Analytics


Jantz Analytics Services

Forecasting Models

The JA-modeling of domestic and international REIT prices is based on multivariate econometric time-series models, capturing the following “stylized facts” observed in empirical studies:

  • (a) clustering of the volatility (heteroscedasticity) of the REIT log-return series.
  • (b) heavy-tailed (non-Gaussian) distributed log-returns innovations (REIT market shocks).
  • (c) asymmetry in the REIT log-returns distributions.
  • (d) time-varying REIT volatility-of-volatility (vol-of-vol).
  • (e) long-range dependence in the REIT intraday log-return time series.
  • (f) non-Gaussian copulas for REIT log-return innovations.

One step ahead (one day or one minute) forecasts are based on 10,000 Monte Carlo scenarios generated by the JA econometric time-series models encompassing (a)-(f).

Remark: Raw historical (without dynamic asset pricing) forecasts solely based on the history of observed REIT log-returns, are only used in JA forecasting tools as REIT industry suggested benchmarks. A raw historical forecast, while (wrongly) viewed as “model-free” assumes that REIT log-returns are independent and identically distributed, which contradicts the empirical studies on observed REIT log-returns time-series. Secondly, the raw historical forecast places zero probability on the future REIT price losses that were not observed over the sample historical period. Finally, raw historical forecasts cannot be used in the valuation of derivatives and portfolio insurance contracts, as the raw historical forecasts do not conform with Dynamic Asset Pricing Theory, and thus, arbitrage-free derivative pricing of REIT portfolios is not possible.

Back-testing Results

The back-testing results are based on the suggested procedures outlined in A Review of Backtesting and Backtesting Procedures by Sean D. Campbell, Finance and Economics Discussion Series Divisions of Research & Statistics and Monetary Affairs Federal Reserve Board, Washington, D.C.


Horse Race for Portfolio Allocation

JA-dynamic forecast with 10,000 one-step ahead scenarios allows forming optimal portfolios with a variety of risk measures and performance ratios, such as:

  • (a) Markowitz min-variance portfolio
  • (b) Markowitz tangent portfolio
  • (c) Min-tail-risk and tangent portfolios based on Conditional Value at Risk (CVaR, or Expected shortfall)
  • (d) Long-short optimal portfolios, including 130/30 Lo-Patel optimal portfolio and zero investment (momentum) portfolio
  • (e) Black-Litterman optimal portfolios

For all cases (a)-(e), a dynamic performance (horse race) comparison with the existing REITs market indexes is performed.

Risk and Performance

JA employs a variety of risk measures and performance ratios to assess the risk-return characteristics of the REIT optimal portfolios. The chosen risk measures are Value-at-Risk (VaR) and CVaR, while the selected performance ratios are chosen following the suggested list in Cheridito, P.; Kromer, E. (2013). “Reward-Risk Ratios”. Journal of Investment Strategies. 3 (1): 1–16, https://www.risk.net/journal-investment-strategies/2317310/reward-risk-ratios

Tools for Risk Insurance

Based on the JA-dynamic pricing models, the valuation of derivative and portfolio insurance contracts of the REIT optimal portfolios and REIT indexes is provided. An analog of VIX for the REIT market is suggested. REIT intrinsic (operational, business) “time-clock” is described dynamically.

Daily Performance Reports

All analytics, portfolio allocations, valuation of derivative contracts are updated daily.


Years of Related Investment Experience

Dallas Office

Stephen T. Crosson
[email protected]

P: (972) 725-7728

3811 Turtle Creek Blvd.
Suite 980
Dallas, Texas 75219

Plano Office

Jimmy H. Jackson
[email protected]

P: (972) 725-7724

1100 Mira Vista Blvd.
Suite 300
Plano, TX 75093