Top

Financial Markets and Portfolio Management

Published in:

20-01-2017

Algorithmic portfolio choice: lessons from panel survey data

Author: Bernd Scherer

Published in: Financial Markets and Portfolio Management | Issue 1/2017

Activate our intelligent search to find suitable subject content or patents.

search-config

AI-assisted search

Off

Abstract

Automated asset management offerings algorithmically assign risky portfolios to individual investors based on investor characteristics such as age, net income, or self-assessment of risk aversion. Using new German household panel data, we investigate the key household characteristics that drive private asset allocation decisions. This information allows us to assess which set of variables should be included in algorithmic portfolio advice. Using heavily cross-validated classification trees, we find that a combination of household balance sheet variables—describing the ability to take risks (e.g., net wealth)—and household personal characteristics—describing the willingness to take risks (e.g., risk aversion)—best explain the cross-sectional variation in household portfolio choice. Our empirical evidence is in line with models of portfolio choice under decreasing relative risk aversion and fixed investment costs. The results suggest the utility of a more holistic modeling of household characteristics. Including background risks in the form of household leverage not only makes investment sense, but is also the new regulatory reality under MIFID II rules. Robo-advisors are strongly advised to act accordingly.

previous article How does the underlying affect the risk-return profiles of structured products?

next article Can investors benefit from the performance of alternative UCITS funds?

Dont have a licence yet? Then find out more about our products and how to get one now:

Springer Professional "Wirtschaft+Technik"

Online-Abonnement

Mit Springer Professional "Wirtschaft+Technik" erhalten Sie Zugriff auf:

über 102.000 Bücher
über 537 Zeitschriften

aus folgenden Fachgebieten:

Automobil + Motoren
Bauwesen + Immobilien
Business IT + Informatik
Elektrotechnik + Elektronik
Energie + Nachhaltigkeit
Finance + Banking
Management + Führung
Marketing + Vertrieb
Maschinenbau + Werkstoffe
Versicherung + Risiko

Jetzt Wissensvorsprung sichern!

inform now

Springer Professional "Wirtschaft"

Online-Abonnement

Mit Springer Professional "Wirtschaft" erhalten Sie Zugriff auf:

über 67.000 Bücher
über 340 Zeitschriften

aus folgenden Fachgebieten:

Bauwesen + Immobilien
Business IT + Informatik
Finance + Banking
Management + Führung
Marketing + Vertrieb
Versicherung + Risiko

Jetzt Wissensvorsprung sichern!

inform now

My Private Banking (2012) estimates that the market for automated investment advice reached USD 20 billion by the end of 2015 and forecasts it to grow to USD 450 billion by 2020.

See Gollier (2001) for a general discussion of the impact of background risk on portfolio choice.

Financial market nonparticipation, documented early on by Mankiv and Zeldes (1991), is now a stylized fact in empirical finance. It is a somewhat puzzling phenomenon, as it does not sit well with the equity risk premium puzzle (high historical risk premium offered by stocks).

This could raise fairness concerns (distributional justice) and throw some doubt on European Central Bank independence.

Note that introducing human capital (outside wealth representing background risk) in combination with no short constraints can also lead to nonparticipation in financial markets. Let $\theta $ denote the fraction of financial wealth relative to total wealth (financial wealth plus human capital). The sensitivity of changes in the market value of human capital with respect to changes in market returns is given by $\beta $ (imagine this to represent the regression beta of human capital returns on asset market returns). In this case

$$\begin{aligned} w^{*}=\mathop {\arg \max }\limits _w \left( {\theta w\mu -\frac{\lambda }{2}\left[ {w^{2}\theta ^{2}\sigma ^{2}+2\left( {1-\theta } \right) \theta w\beta \sigma ^{2}} \right] } \right) =\left( {\frac{1}{\theta }} \right) \frac{\mu }{\lambda \sigma ^{2}}+\left( {1-\frac{1}{\theta }} \right) \beta \end{aligned}$$

Unconstrained optimal portfolio weights become negative if

$$\begin{aligned} \frac{\mu }{\lambda \sigma ^{2}}<\beta \left( {1-\theta } \right) . \end{aligned}$$

This occurs when human capital contains a large beta component and simultaneously assumes a large fraction of total wealth, that is, the household holds too much equity in its human capital. Combined with a long-only constraint, this leads to nonparticipation (zero portfolio weight).

Investment robots can reduce the number of nonparticipating households by lowering $\mu $ (cheap ETF-based access to diversified beta) or by reducing $\phi $ (individualized portfolio advice reduces complexity and monitoring costs).

Eurosystem Household Finance and Consumption Survey (HFCS) as described in European Central Bank (2012).

For ease of replication we use the same data identifiers as described in the HFCS core and noncore variables catalogue provided by ECB (2012a, 2012b). We collect personal information on risk aversion (“HD1800”), age (“RA0300”), gender (“RA0200”), marital status (“PA0100”), household members, and education (“PA0200”) for each household (given by its unique household ID). Financial wealth consists of cash (“HD1110”), savings accounts (“HD1210”), mutual funds (“HD1320g”), bonds (“HD1420”), and shares (“HD1510”). For mutual fund holdings we have the additional information about whether they are mostly equity (“HD1320a”). Net wealth (“DN3001”) is available as an already consolidated figure, that is, we do not need to calculate it. We collect income (“DI2000”) and real estate ownership (“DA1100,” “DA1120”). Using outstanding mortgage (“DL1100”) allows us to calculate home leverage. Human capital is given as after-tax disposable income multiplied by the number of years left until an assumed retirement age of 65.

Curcuru et al. (2004) provide a set of variables that could explain heterogeneities in observed portfolios.

The classic reference is Hastie et al. (2001).

All calculations are performed using ctree() from the R package rparty.

Breiman, L., Friedman, J., Stone, C., Olshen, R.: Classification and Regression Trees. Wadsworth Statistics, Wadsworth (1984)

Campbell, J. Y.: Household Finance. J. Financ. , 61(4), 1553–1604 (2006)

Curcuru, S., Heaton, J., Lucas, D., Moore, D.: Heterogeneity and portfolio choice: theory and evidence. In: Yacine, A., Hansen, L. (eds.), Handbook of Financial Econometrics: Tools and Techniques, pp. 337–382 Elsevier (2004)CrossRef

European Central Bank. HFCS Core Variables Catalogue (2012)

Frantantoni, M.: Home-ownership and investment in risky assets. J. Urb. Econ. 44(1), 27–42 (1998)CrossRef

Gollier, C.: The Economics of Risk and Time. Massachusetts Institute of Technology, Massachusetts (2001)CrossRef

Grinblatt, M., Keloharju, M., Linnaunmaa, J.: IQ and stock market participation. J. Financ. 66(6), 2121–2164 (2011)CrossRef

Guiso, L., Halliassos, M., Jappelli, T.: Household stockholding in Europe: where do we stand and where do we go? Econ. Policy 18(36), 123–170 (2003)CrossRef

Hastie, T., Tishrani, R., Friedman, J.: The Elements of Statistical Learning: Data Mining. Inference and Prediction. Springer, New York (2001)CrossRef

Heaton, J., Lucas, D.: Stock prices and fundamentals. NBER/Macroecon. Ann. (MIT Press) 14(1), 213–242 (2000)

Hsu, C.: What drives equity market non-participation? N. Am. J. Econ. Financ. 23, 86–114 (2012)CrossRef

King, M., Leape, J.: Asset accumulation, information, and the life cycle. NBER Working Paper No. 2392 (1987)

Mankiv, N., Zeldes, S.: The consumption of stockholders and non-stockholders. J. Financ. Econ. 29(1), 97–112 (1991)CrossRef

My Private Banking: Robo-advisors 2.0: how automated investing is infiltrating the wealth management industry. Research Report (2012)

Scherer, B: Asset allocation and divorce risk. In: Rudd, A., Satchell, S. (eds.), Quantitative Approaches to High Net Worth Investment, pp. 269–280 (2014)

Vissing-Jorgenson, A.: Towards an explanation of household portfolio choice heterogeneity: non-financial income and participation cost structures. NBER Working Paper No. 8884 (2002)

Title: Algorithmic portfolio choice: lessons from panel survey data
Author: Bernd Scherer
Publication date: 20-01-2017
Publisher: Springer US
Published in: Financial Markets and Portfolio Management / Issue 1/2017
Print ISSN: 1934-4554
Electronic ISSN: 2373-8529
DOI: https://doi.org/10.1007/s11408-016-0282-8

Springer Professional

Abstract

Please log in to get access to your license.

Dont have a licence yet? Then find out more about our products and how to get one now:

Springer Professional "Wirtschaft+Technik"

Springer Professional "Wirtschaft"

Other articles of this Issue 1/2017

How does the underlying affect the risk-return profiles of structured products?

Turan G. Bali, Yigit Atilgan, and K. Ozgur Demirtas: Investing in hedge funds: a guide to measuring risk and return characteristics

Can investors benefit from the performance of alternative UCITS funds?

A good pair: alternative pairs-trading strategies