Different Measures Of Risk Research Articles

From the Editors

From the Editors

A Comparison of Empirically Based and Structured Professional Judgment Estimation of Risk Using the Structured Assessment of Violence Risk in Youth

This study builds on a long-standing debate focusing on whether structured professional judgment (SPJ) or empirically based methods of risk estimation are more valid and reliable measures of future behavior by comparing three different measures of risk. Data were collected from the Structured Assessment of Violence Risk in Youth administered to a sample of 177 adjudicated juvenile offenders prior to being placed on probation. Three measures of risk were examined: an empirically derived measure of risk using latent class analysis, a violence risk based on SPJ, and a nonviolent delinquency risk based on SPJ. The ability of each measure to predict probation-related outcomes and recidivism was also addressed. Results provide moderate support for the continued use of the SPJ framework and highlight the need for future research regarding risk assessment procedures in juvenile justice settings.

Low-Risk Anomalies in Global Fixed Income: Evidence from Major Broad Markets

In this paper we present the most compelling empirical evidence yet of a low-risk anomaly in fixed income markets. We show that portfolios invested in bonds with the lowest risk would have delivered the largest positive alpha and highest Sharpe ratios and portfolios invested in riskier bonds would have delivered the most negative alpha and lowest Sharpe ratios. Our results proved extremely robust and were confirmed for government bonds, quasi-government and foreign government bonds, securitized and collateralized bonds, corporate investment grade bonds, corporate high yield bonds, emerging market bonds and aggregates of some of these universes. We considered bonds denominated in USD, EUR, GBP and JPY separately and the results proved invariant to the currency. We confirmed the robustness of the results by using different measures of risk. The results were produced using data from the Bank of America Merrill Lynch database from January 1997 which includes 85,442 individual bonds in the 192 months analyzed.

Cognitive vulnerability to depression: A comparison of the weakest link, keystone and additive models

Multiple theories of cognitive vulnerability to depression have been proposed, each focusing on different aspects of negative cognition and utilising different measures of risk. Various methods of integrating such multiple indices of risk have been examined in the literature, and each demonstrates some promise. Yet little is known about the interrelations among these methods, or their incremental validity in predicting changes in depression. The present study compared three integrative models of cognitive vulnerability: the additive, weakest link, and keystone models. Support was found for each model as predictive of depression over time, but only the weakest link model demonstrated incremental utility in predicting changes in depression over the other models. We also explore the correlation between these models and each model's unique contribution to predicting onset of depressive symptoms.

Is the stock market efficient in bad times and inefficient in good times?

This article examines the cross sectional predictability of stock returns within the framework of time varying risk premia and asymmetric risk. In bad times, small, value and cyclical stocks are riskier than big, growth and noncyclical stocks, thereby explaining the value, size and cyclical premia. In contrast, in good times, value, big and noncyclical stocks generate higher average returns than do growth, small and cyclical stocks, despite the fact that they are not riskier. Furthermore, empirical tests of macroeconomic models highlight that average returns are commensurate with risk only in bad times. These results are robust to different measures of risk and different sets of test assets.

Different Risk-Adjusted Fund Performance Measures: A Comparison

Traditional risk-adjusted performance measures, such as the Sharpe ratio, the Treynor index or Jensen's alpha, based on the mean-variance framework, are widely used to rank mutual funds. However, performance measures that consider risk by taking into account only losses, such as Value-at-Risk (VaR), would be more appropriate. Standard VaR assumes that returns are normally distributed, though they usually present skewness and kurtosis. In this paper we compare these different measures of risk: traditional ones vs. ones that take into account fat tails and asymmetry, such as those based on the Cornish-Fisher expansion and on the extreme value theory. Moreover, we construct a performance index similar to the Sharpe ratio using these VaR-based risk measures. We then use these measures to compare the rating of a set of mutual funds, assessing the different measures' usefulness under the Basel II risk management framework.

Constructing Risk Measures from Uncertainty Sets

We illustrate the correspondence between uncertainty sets in robust optimization and some popular risk measures in finance and show how robust optimization can be used to generalize the concepts of these risk measures. We also show that by using properly defined uncertainty sets in robust optimization models, one can construct coherent risk measures and address the issue of the computational tractability of the resulting formulations. Our results have implications for efficient portfolio optimization under different measures of risk.

Who drops‐out? Do measures of risk to self and to others predict unplanned endings in primary care counselling?

AbstractAims: Unplanned endings, where clients unilaterally end therapy, are of concern for psychological therapy services generally as they raise questions about the appropriateness of the treatment and it's delivery for some clients. Limited available data indicates that those who drop‐out often have more severe symptoms at entry, and have poorer clinical outcomes. This raises further questions about risk to self and others for those clients who leave therapy prematurely and how these clients might be identified and kept engaged. Method: This paper uses a large dataset of CORE data collected routinely in a primary care counselling service between 2000 and 2003. Logistic regression was utilised to consider different measures of risk and other client characteristics recorded at assessment to predict drop‐out from the service. Results: These indicate that younger age, greater psychological distress at assessment, an addiction problem and greater risk to others, are associated with an unplanned ending. However, no reliable logistic regression model could be produced. This may be partly due to data quality issues or important characteristics not being available in the data. Implications for practice: The paper concludes that counsellors should actively seek to minimise unplanned endings, as amongst them may be represented the more distressed and risky clients referred to primary care counselling.

Adding an Alcohol‐Related Risk Score to an Existing Categorical Risk Classification for Older Adults: Sensitivity to Group Differences

To evaluate a new alcohol-related risk score for research use. Using data from a previously reported trial of a screening and education system for older adults (Computerized Alcohol-Related Problems Survey), secondary analyses were conducted comparing the ability of two different measures of risk to detect post-intervention group differences: the original categorical outcome measure and a new, finely grained quantitative risk score based on the same research-based risk factors. Three primary care group practices in southern California. Six hundred sixty-five patients aged 65 and older. A previously calculated, three-level categorical classification of alcohol-related risk and a newly developed quantitative risk score. Mean post-intervention risk scores differed between the three experimental conditions: usual care, patient report, and combined report (P<.001). The difference between the combined report and usual care was significant (P<.001) and directly proportional to baseline risk. The three-level risk classification did not reveal approximately 57.3% of the intervention effect detected by the risk score. The risk score also was sufficiently sensitive to detect the intervention effect within the subset of hypertensive patients (n=112; P=.001). As an outcome measure in intervention trials, the finely grained risk score is more sensitive than the trinary risk classification. The additional clinical value of the risk score relative to the categorical measure needs to be determined.

The Effect of Health Risk on Housing Values: Evidence from a Cancer Cluster.

This paper measures the impact of an outbreak of pediatric leukemia on local housing values. A model of location choice is used to describe conditions under which the gradient of the hedonic price function with respect to pediatric leukemia risk is equal to household marginal willingness to pay to avoid risk. This equalizing differential is estimated using property-level sales records from a county in Nevada where residents recently experienced a severe increase in pediatric leukemia. Housing prices are compared before and after the increase with a nearby county acting as a control group. The variation in health risk over time makes it possible to control for unobserved differences across locations. In addition, because many houses were sold repeatedly during the sample period it is possible to control for property-specific heterogeneity. The results indicate that housing values decreased 15.6 percent during the period of maximum risk. Results are similar for different measures of risk and across houses of different sizes. Using lifetime estimates of risk derived from a Bayesian learning process the results imply that the statistical value of pediatric leukemia is $5.6 million. These estimates provide some of the first market-based estimates of the value of health for children.

The Diversity of Risk Among Age-62 Retired Worker Beneficiaries

This article focuses on age-62 retired-worker beneficiaries, a group whose well-being may be affected by changes in Social Security retirement ages. The analysis: 1) develops different measures of risk of a poor retirement; 2) applies these measures to developing a range of estimates of the risk for age-62 beneficiaries at the threshold of retirement; and 3) assesses how the circumstances of and risks experienced by these beneficiaries vary by such factors as race, gender, health status and marital status. The findings point to great diversity of circumstances among these early retirees and suggest that narrow conceptions of risk may fall short of fully identifying the distributive consequences of retirement age changes, especially for African Americans, Hispanics, low-income, unmarried individuals and unhealthy early retirees.

Reconciling different measures of risk in the treatment of hypertension: a community-based study.

There is disagreement as to how cardiovascular risk in hypertensive patients should be measured. In particular, whether absolute risk measurement alone is a realistic index on which to base treatment goals. A cross-sectional study on 895 treated hypertensive patients in 18 general practices in the UK. Reporting on the distribution and magnitude of age-specific absolute risk. The percentage of individuals with controlled hypertension whose absolute risk exceeds their age/sex absolute risk standard and the percentage of individuals with uncontrolled hypertension whose risk is less than their age/sex absolute risk standard. Overall, 62.8% (95% CI 59.6-66%) individuals had an absolute risk that exceeded 20% over 10 years. The magnitude of absolute risk was considerable (range 3.5-87.8%) and increasing absolute risk was significantly associated with age. Of those individuals with controlled hypertension 50.5% (95% CI 45.1-55.8%) had an absolute risk which exceeded their age/sex absolute risk standard. Conversely, 30.4% (95% CI 26.6-34.2%) of those with uncontrolled hypertension had an absolute risk that was less than their age/sex absolute risk standard. The distribution and magnitude of absolute risk is significantly associated with age. Appreciation of such a relationship is needed when setting realistic treatment goals according to an absolute risk standard, particularly in the elderly. In addition, the use an age/sex absolute risk standard is likely to further modify treatment goals in individuals at high and low absolute risk of cardiovascular disease.

Measuring risk taking

Risk taking can be defined as the preparedness of a student to attempt to answer a question when not certain of the result. In this paper, three different measures of risk taking were applied to multichoce paper. Specifically, data from the Australian Mathematics Competition, one of the largest data bases of its type, for the years 1988 and 1989, are analysed to measure risk taking by groups of students classified by gender, school year, and achievement level.

The impact of international listings on risk: Implications for capital market integration

In this paper, we examine the impact of international listing on common-stock risk. While previous research has used event study methodology, our research focuses on permanent shifts in risk. Different measures of risk are estimated to test for intertemporal shifts in risk attributable to an overseas listing. No significant shifts in risk from international listing are documented. The results are robust with respect to the location and year of listing. These findings suggest that: (1) markets are already reasonably well integrated; or (2) listing is an ineffective mechanism for reducing segmentation.

AN ANALYSIS OF ALTERNATIVE MEASURES OF INVESTMENT RISK

DESPITE THE THEORETICAL and empirical difficulties in defining and measuring risk, the community is concerned with the investment of alternative securities and firms. This study restricts itself to an analysis of the similarity between two widely employed measures of risk: common stock systematic risk and corporate bond ratings.' The capital asset pricing model [11, 14, 21] specifies that in a properly diversified portfolio, the relevant measure of risk is the covariance of the asset return with the market return divided by the variance of the market return-known as systematic risk or beta. In an efficient market an investor is rewarded only for bearing systematic (i.e., market-wide or non-diversifiable) risk. Systematic risk information is now routinely provided investors by many advisory agencies on a large number of common stocks. A number of recent studies [2, 4, 7] have indicated varying relationships between systematic risk and financialor firm-related estimates of risk. Corporate bond ratings provide a simple indication of the relative risk associated with the fixed income offerings of a large number of publicly owned firms. These ratings, which attempt to look at 'worst' potentialities in the 'visible' future. . [13, p.v.], appear to be based on a combination of objective and subjective factors [16]. Earlier studies have indicated a consistent relationship between bond ratings and actual rates of default [1, 9]. In recent years a number of models for predicting ratings, incorporating a combination of bondrelated and financialor firm-related variables, have been developed [10, 15, 17, 22]. While bond rating agencies do not claim complete precision, the ratings seek to provide the American investor with a simple system of gradation by which relative qualities of bonds may be noted [13, p.v.]. The purpose of this study is to examine the relationship between these two different measures of risk that are readily available to virtually all investors. Given the high level of efficiency in our capital markets [5], and the fact that previous studies of both common stock systematic risk and bond ratings have indicated definite relationships with certain financial variables, we hypothesize that common stock sys-

Mathematical Model for Flood Risk Evaluation

A mathematical model is developed for use with partial-duration data that relates design flow to several different measures of risk. The analysis is based on the Poisson probability law and the probability theory of sums of a random number of random variables. Finally, a statistical procedure is presented for estimating design equation parameters from a partial-duration series of data. A graphical procedure for solving these equations is also presented.