Nico L. van der Sar

Household Cost Functions and Equivalence Scales

In this paper we describe a simple method to estimate household cost functions and family equivalence scales. It is an alternative to standard methods as it does not assume strong postulates about utility maximization nor any functionally specified model equations. The data requirements are extremely modest. We assume interpersonal ordinal comparability in the sense of Sen (1976). Empirical evidence for eight European countries and the U.S.A. shows the feasibility of the method and the stability of its results.

Individual expectations, risk perception and preferences in relation to investment decision making☆

Abstract Both expectations and preferences with respect to stock returns are put into a model which is estimated for a sample drawn from the Dutch Central Union of Investment Study Clubs in 1987. Our approach differs from most economic research since it makes use of economic-psychological variables concerning future developments of financial and macro-economic factors, as they are expected by the respondents, multiplied by their perceived importance weights. Also, the Pratt-Arrow relative risk aversion measure is applied to estimate the dependence of the return-risk attitude on investor-specific variables. Finally, some conclusions are drawn with respect to investment decision making.

CALENDAR EFFECTS ON THE AMSTERDAM STOCK EXCHANGE

For daily data on a value-weighted index of all shares in the Netherlands (1981-1998), we find abnormally high returns in the pre-Christmas period of the second half of December and around the turn of the month, whereas returns are negative and volatility is relatively high on the Mondays where the previous weeks return is below zero. Furthermore, our evidence indicates the presence of an ARCH(1) effect. Our intraday results based on an equal-weighted index of the leading shares in the Netherlands (1986-1993) reveal a U-shaped return pattern over each trading day, Monday morning excluded. Moreover, our tests reveal repeated price adjustments mostly in the same direction around and over non-trading periods. We argue that the arrival of private information is affected by strategic and behavioral factors incompatible with the risk-return paradigm.For daily data on a value-weighted index of all shares in the Netherlands (1981-1998), we find abnormally high returns in the pre-Christmas period of the second half of December and around the turn of the month, whereas returns are negative and volatility is relatively high on the Mondays where the previous weeks return is below zero. Furthermore, our evidence indicates the presence of an ARCH(1) effect. Our intraday results based on an equal-weighted index of the leading shares in the Netherlands (1986-1993) reveal a U-shaped return pattern over each trading day, Monday morning excluded. Moreover, our tests reveal repeated price adjustments mostly in the same direction around and over non-trading periods. We argue that the arrival of private information is affected by strategic and behavioral factors incompatible with the risk-return paradigm.

The evaluation question approach: A method of measuring attitudes

Abstract This paper presents the Evaluation Question Approach (EQA), which can measure attitudes to social and economic issues within a cognitive-evaluative framework. The basic idea behind the EQA is that verbal evaluation labels are offered to a respondent who then associates a description or manifestation of the object in question. Instead of supplying stimuli for corresponding sensations, as is the case using psychophysical methods, the approach being examined does the reverse. It may be seen as the opposite of the vignette technique with which a methodoligically-based comparison is made. Verbal labels are studied to see if they are otherwise interpreted by different respondents. This important scaling problem is applied to a 1983 Boston (USA) data-set of about 500 respondents. The result of this empirical test supports the hypothesis that there are no interpretation differences. In conclusion, the EQA appears to be a viable general method and can be applied to a variety of subjects in all behavioral sciences.

Volatility, Investor Uncertainty, and Dispersion

This paper examines the use of survey-based measures in volatility forecasting. We argue that the dispersion of individual mean return forecasts bridges the gap between individual volatilities and aggregate volatility. We use data coming from a repeated survey to capture volatility and mean return expectations of investors, and to produce aggregate volatility forecasts. Our survey-based measures are consistent and quantitatively similar with forecasts based on GARCH and implied volatility models. This result is robust to both in-sample and out-of-sample comparisons and in response to news. As an implication, disagreement can be regarded as an integral part of risk.

Social Distance on the Income Dimension

Social justice refers, according to Jasso (1980), to a distribution of goods. As long as we may assume that all citizens have the same distribution in mind, we may concentrate on the definition of a social justice index. Preliminary to that problem is the question whether all citizens perceive the same distribution of goods. In this chapter, we show that different people at different levels of society will have different perceptions of the distribution. This is due to the fact that there exists a varying social distance between citizens.

Utility of Wealth and Relative Risk Aversion: Operationalization and Estimation

Next to expectations, preferences play an important part in explaining individual investment decision making. In contrast with the case of expectations, most financial models do not presuppose homogeneity with respect to preferences, but leave room for variation between individuals. In this article we employ the social filter theory to model the concept of utility. The investor group on which we focus consists of a sample drawn from the members of the Dutch Central Union of Investment Study Clubs. By means of verbal choice behavior we operationalize and determine empirically the utility function of wealth as a function of club-specific characteristics, and we analyze the implications for the corresponding relative risk aversion.

ASSET ALLOCATION AND THE INVESTOR'S RELATIVE RISK AVERSION

This paper addresses the issue of how an investor allocates his wealth among assets and examines the nature of the dependency of the portfolio selection on the willingness to take on extra risks. We focus on two two-asset allocation models where only the relative risk aversion is needed to establish the investor’s risk-return trade-off. A methodological comparison is made between Sharpe’s (1987) analysis concerning the two-asset allocation problem and the mean-variance approach, based on the second order Taylor series approximation.

Social stratification: Individual importance and social welfare

A method to estimate a households social status in a stratified society is described. We do not leave it up to the respondents what is to be understood by status, as most methods do, but define it as the weight of the household with respect to social welfare. In order to operationalize this individual relative importance concept we measure and utilize the individual welfare evaluations. Our empirical results for a data set collected in the Boston area in 1983 show the dependency of social status and social stratification on individual characteristics. A comparison made with some other methods shows the feasibility of our approach.