The Market Timing Power of Moving Averages: Evidence from US REITs and REIT Indexes

• Published date: 01 June 2014
• DOI: http://doi.org/10.1111/irfi.12018
• Author: Paskalis Glabadanidis
• Date: 01 June 2014

The Market Timing Power of

Moving Averages: Evidence from US

REITs and REIT Indexes*

PASKALIS GLABADANIDIS

Finance Discipline, University of Adelaide Business School, Adelaide, Australia

ABSTRACT

I present evidence that a moving average (MA) trading strategy dominates

buying and holding the underlying asset in a mean-variance sense using

monthly returns of value-weighted and equal-weighted US REIT indexes over

the period January 1980 until December 2010. The abnormal returns are

largely insensitive to the four Carhart factors and produce economically and

statistically significant alphas of between 10 and 15% per year after transaction

costs. This performance is robust to different lags of the MA and in subperiods

while investor sentiment, liquidity risks, business cycles, up and down

markets, and the default spread cannot fully account for its performance. The

MA strategy works just as well with randomly generated returns and

bootstrapped returns. The substantial market timing ability of the MA strategy

appears to be the main driver of the abnormal returns. The returns to the MA

strategy resemble the returns of an imperfect at-the-money protective put

strategy relative to the underlying portfolio. The lagged signal to switch has

substantial predictive power over the subsequent return of the REIT index. The

MA strategy avoids the sharp downturn at the beginning of 2008 and substan-

tially outperforms the cumulative returns of the buy-and-hold strategy using

all of the 20 REIT indexes. The results from applying the MA strategy with 274

individual REITs largely corroborate the findings for the REIT indexes.

JEL Classification: G11, G12, G14.

I. INTRODUCTION

Technical analysis involves the use of past and current market price, trading

volume and, potentially, other publicly available information to try and predict

future market prices. It is highly popular in practice with plentiful financial

trading advice that is based largely, if not exclusively, on technical indicators. In

a perhaps belated testament to this fact, consider the following quote from the

New York Times issue dated March 11, 1988: ‘Starting today the New York Times

will publish a comprehensive three-column market chart every Saturday . . .

History has shown that when the S&P index rises decisively above its (moving)

average the market is likely to continue on an upward trend. When it is below

* I would like to thank Huining Cao, Sudipto Dasgupta, Dogan Tirtiroglu, Takeshi Yamada, and an

anonymous referee for their helpful comments and suggestions. Any remaining errors are my own.

bs_bs_banner

International Review of Finance, 14:2, 2014: pp. 161–202

DOI: 10.1111/irfi.12018

© 2013 International Review of Finance Ltd. 2013

the average that is a bearish signal’. The evidence reported in Brock et al. (1992)

points toward a significant predictive ability of some technical indicators. A

more thorough study of a large set of technical indicators by Lo et al. (2000) also

found some predictive ability especially when moving averages (MAs) are con-

cerned. Zhu and Zhou (2009) provide a solid theoretical reason why technical

indicators could be a potentially useful state variable in an environment where

investors need to learn over time the fundamental value of the risky asset they

invest in. More recently, Neely et al. (2010, 2011) find that technical analysis

has as much forecasting power over the equity risk premium as the information

provided by economic fundamentals. The practitioners literature also includes

Faber (2007) and Kilgallen (2012) who thoroughly document the risk-adjusted

returns to the MA strategy using various portfolios, commodities, and curren-

cies. In a similar spirit, Glabadanidis (2012) shows the power of the MA switch-

ing strategy at the monthly frequency for various sets of portfolios sorted on

size, book-to-market ratio, earnings and dividend yield, past price reversals and

momentum, as well as for industry portfolios.

The main findings of this study are as follows. First, I present evidence that

the returns to a simple MA switching strategy dominate in a mean-variance

sense the returns to a buy-and-hold strategy of the underlying US REIT index.

Second, I demonstrate that there is relatively infrequent trading with fairly long

periods when the MA strategy is also holding the underlying assets and the

break-even transaction costs (BETCs) are on the order of 5–7% per transaction.

Third, even though there is overwhelming evidence of market timing ability of

the MA switching strategy, cross-sectional differences remain between the port-

folio abnormal returns and several alphas are still statistically significant. These

differences persist when controlling for the four-factor Carhart (1997) model for

portfolios formed on past price returns. Fourth, conditional models explain to

a certain degree the MA abnormal returns but do not completely eliminate the

significant alphas. Fifth, I offer a rationale for the success of the MA strategy in

outperforming a buy-and-hold strategy by noting that the strategy payoffs

resemble an imperfect protective put. Sixth, I show that the lagged indicator to

switch into the risky asset has substantial predictive ability over subsequent

index returns over and above the predictability contained in standard instru-

mental variables, like the default spread, investor sentiment, recession dummy

variable, and liquidity risk. Last but not least, the strategy is robust to randomly

generated stock returns and bootstrapped historical returns.

This paper is similar in spirit to Han et al. (2012). However, a few important

differences need to be pointed out. First, I use monthly value-weighted and

equal-weighted returns of 20 US REIT indexes. The reason for investigating

REITs is that they provide valuations for otherwise illiquid and highly location-

specific as well as purpose-specific assets. Value-weighted portfolios at a

monthly frequency should have a much smaller amount of trading going on

inside the portfolio compared to the daily equal-weighted portfolios investi-

gated by Han et al. (2012). Second, I make a case for using monthly returns

rather than daily returns in order to reduce the frequency of trading and achieve

International Review of Finance

162 © 2013 International Review of Finance Ltd. 2013

much higher BETCs. The highlights of this study are the extremely good

performance of the MA portfolios relative to buying and holding (BH) the

underlying portfolios, the infrequency of trading, and the very large BETCs.

This paper proceeds as follows. Section II presents the MA investment strat-

egy. Section III presents evidence regarding the profitability of the MA switching

strategy. Section IV investigates the robustness of the results in a number of

ways. Section V discusses the potential drivers of the performance of the MA

strategy over the business cycle and controls for sensitivity to up or down

markets, investor sentiment, the default premium, and an aggregate liquidity

factor. Sections VI and VII explore the predictive power of the MA signal to

switch over future REIT index returns and the performance of the MA strategy

using international stock markets and US Exchange Traded Funds (ETF) as well

as its performance at the daily return frequency. Section VIII offers a few

concluding remarks and discusses potential areas of future research.

II. MA TIMING STRATEGIES

I use monthly value-weighted1returns of two sets of 10 US REIT indexes. The

data are readily available from the CRSP/Ziman Data set. The sample period

starts in January 1980 and ends in December 2010.

The following exposition of the MA strategy follows closely the presentation

in Han et al. (2012). Let Rjt be the return on portfolio jat the end of month tand

let Pjt be the respective price level of that portfolio. Define the MA of portfolio

jA

jt,Lat time twith length Lperiods as follows:

APP PP

L

jt L

jt L jt L jt jt

,.=++++

−+ −+ −12 1

(1)

Throughout most of the paper, I use a MA of length L=24 months. Later on, in

the robustness checks I also present results for all sets of portfolios with lags of 6,

12, 36, 48, and 60 months. According to Brock et al. (1992), the MA in its various

implementations is the most popular strategy followed by investors who use

technical analysis. The way I implement the MA strategy in this paper is to

compare the closing price Pjt at the end of every month to the running MA Ajt,L.

If the price is above the MA this triggers a signal to invest (or stay invested if

already invested at t−1) in the portfolio in the next month t+1. If the price is

below the MA this triggers a signal to leave the risky portfolio (or stay invested in

cash if not invested at t−1) in the following month t+1. As a proxy for the

risk-free rate, I use the returns on the 30-day US Treasury Bill.

More formally, the returns of the MA switching strategy can be expressed as

1 I use value-weighted portfolio returns to limit the amount of trading inside the various

portfolios. This provides an important control case when compared with the results for

equal-weighted portfolio returns. However, this may understate the break-even transaction

costs as equal-weighted portfolios require a lot of trading to be replicated. Similarly, I use

monthly returns in order to have a trading strategy that trades less frequently. The results with

daily portfolio returns are similar and lead to higher abnormal returns without a dispropor-

tionately more frequent trading.

Market Timing Power of Moving Averages

163© 2013 International Review of Finance Ltd. 2013