Australian stock indexes and the four-factor model

Stock indexes are passive 'value-weighted' portfolios and should not have alphas which are significantly different from zero. If an index produces an insignificant alphan, then significant alphas for equity funds using this index can be attributed solely to manager performance. However, recent literature sugests that US Stock indexes can demonstrate significant alphas, which ultimately raise questions regarding equity fund manager performance in both the US and abroad. in this paper, we employ the Carhart four-factor model and newly available Asian-Pacific risk factors to generate alphas and risk factor loadings for eight Australian stock indexes from January 2004 to December 2012. We ifnd that the initial full sample period analysis does not provide indication of significant alphas in the indexes examined. However, by carrying out 36-month rolling regressions, we discover at least four significant alphas in seven of the eight indexes and factor loading variability. As previously reported in the US, this paper confirms similar issues with the four-factor model using Australian stock indexes and performance benchmarking. In effectively measuring Australian equity fund manager performance, it is therefore essential to evaluate a fund's alpha and risk factors relative to the alpha and risk factors of the appropriate benchmark index.

Confirmatory factory analysis of the Neck Disability Index in a general problematic neck population indicates a one-factor model

BACKGROUND CONTEXT: The Neck Disability Index frequently is used to measure outcomes of the neck. The statistical rigor of the Neck Disability Index has been assessed with conflicting outcomes. To date, Confirmatory Factor Analysis of the Neck Disability Index has not been reported for a suitably large population study. Because the Neck Disability Index is not a condition-specific measure of neck function, initial Confirmatory Factor Analysis should consider problematic neck patients as a homogenous group. PURPOSE: We sought to analyze the factor structure of the Neck Disability Index through Confirmatory Factor Analysis in a symptomatic, homogeneous, neck population, with respect to pooled populations and gender subgroups. STUDY DESIGN: This was a secondary analysis of pooled data. PATIENT SAMPLE: A total of 1,278 symptomatic neck patients (67.5% female, median age 41 years), 803 nonspecific and 475 with whiplash-associated disorder. OUTCOME MEASURES: The Neck Disability Index was used to measure outcomes. METHODS: We analyzed pooled baseline data from six independent studies of patients with neck problems who completed Neck Disability Index questionnaires at baseline. The Confirmatory Factor Analysis was considered in three scenarios: the full sample and separate sexes. Models were compared empirically for best fit. RESULTS: Two-factor models have good psychometric properties across both the pooled and sex subgroups. However, according to these analyses, the one-factor solution is preferable from both a statistical perspective and parsimony. The two-factor model was close to significant for the male subgroup (p<.07) where questions separated into constructs of mental function (pain, reading headaches and concentration) and physical function (personal care, lifting, work, driving, sleep, and recreation). CONCLUSIONS: The Neck Disability Index demonstrated a one-factor structure when analyzed by Confirmatory Factor Analysis in a pooled, homogenous sample of neck problem patients. However, a two-factor model did approach significance for male subjects where questions separated into constructs of mental and physical function. Further investigations in different conditions, subgroup and sex-specific populations are warranted.

A factor model of seasonality in stock returns

Published as an article in: The Quarterly Review of Economics and Finance, 2004, vol. 44, issue 2, pages 224-236.

Essays on Applications of the Factor Model

Thesis (Ph.D.)--University of Washington, 2013

A five factor model of grief : a q-methodological study /

Q-methodology permitted 41 people to communicate their perspective of grief. In an attempt to clarify the research to date and to allow those who have experienced this human journey to direct the scientists, 80 statements were chosen to present to the participants based on the research from academic and counselling sources. Five different perspectives emerged from the Q-sorts and factor analysis. Each perspective was valuable for the understanding of different groups of mourners. They were interpreted using questionnaire data and interview information. They are as follows: Factor 1- Growth Optimism; Factor 2 - Schema Destruction and Negative Affect; Factor 3- Identification with the Deceased Person; Factor 4- Intact World view with High Clarity and High Social Support; Factor 5- Schema Destruction with High Preoccupation and Attention to Emotion. Some people grow in the face of grief, others hold on to essentially the same schemas and others are devastated by their loss. The different perspectives reported herein supply clues to the sources of these differing outcomes. From examination of Factor 1, it appears that a healthy living relationship helps substantially in the event of loss. An orientation toward emotions that encourages clarity, exemplified by Factor 4, without hyper-vigilance to emotion may be helpful as well. Strategies for maintaining schematic representations of the world with little alteration include: identification with the values of the deceased person, as in Factor 3 and reliance on social support and/or God as demonstrated by Factor 4. When the relationship had painful periods, social support may be accessed to benefit some mourners. When the person's frame of reference or higher order schemas are assaulted by the events of loss, the people most at risk for traumatic grief seem to be those with difficult relationships as indicated by Factor 5 individuals. When low social support, high attention to emotion with low clarity and little belief that feelings can be altered for the better are also attributes of the mourner devastating grief can result. In the end, there are groups of people who are forced to endure the entire process of schema destruction and devastation. Some appear to recover in part and others appear to stay in a form of purgatory for many years. The results of this study suggest that, those who experience devastating grief may be in the minority. In the future interventions could be more specifically addressed if these perspectives are replicated in a larger, more detailed study.

A Semi-Parametric Factor Model for Interest Rates.

Understanding the dynamics of interest rates and the term structure has important implications for issues as diverse as real economic activity, monetary policy, pricing of interest rate derivative securities and public debt financing. Our paper follows a longstanding tradition of using factor models of interest rates but proposes a semi-parametric procedure to model interest rates.

The use of the bi-factor model to test the uni-dimensionality of a battery of reasoning tests

Resumen tomado de la publicación

Real-time factor model forecasting and the effects of instability

Factor forecasting models are shown to deliver real-time gains over autoregressive models for US real activity variables during the recent period, but are less successful for nominal variables. The gains are largely due to the Financial Crisis period, and are primarily at the shortest (one quarter ahead) horizon. Excluding the pre-Great Moderation years from the factor forecasting model estimation period (but not from the data used to extract factors) results in a marked fillip in factor model forecast accuracy, but does the same for the AR model forecasts. The relative performance of the factor models compared to the AR models is largely unaffected by whether the exercise is in real time or is pseudo out-of-sample.

The role of personality in leadership: an application of the five-factor model in the Australian military

This study investigated the relation between the five-factor model (FFM) of personality trait domains and leadership effectiveness. Ninety-nine Australian Army commissioned officers completed the NEO Personality Inventory-Revised (NEO-PI-R; Costa & McCrae, 1992) and were rated by their superior officer on the Australian Army annual leadership effectiveness evaluation schedule. Participants indicated whether they had been selected to attend a leadership promotion course at the Army Command and Staff College, widely regarded within the Army as indicative of an officer's effectiveness. It was hypothesized that leadership effectiveness would be predicted by the personality trait domains of high Conscientiousness, Openness, Agreeableness, and Extraversion and by low Neuroticism. High Conscientiousness and low Extraversion scores predicted high leadership effectiveness and the likelihood of attending the leadership promotion course. High Openness scores also predicted the likelihood of attending the promotion course. The results support the utility of the FFM in exploring the role of personality in leadership effectiveness among military leaders.

Longitudinal evaluation of a risk-factor model for adolescent suicidality

This research developed two best-fitting structural equation models of risk factors for adolescent depression and suicidality: a core model, which included parenting factors, gender, depression, and suicidality, and an extended model, which also encompassed personality traits (Introversion and Impulsivity) and mood factors (Anxiety and Anger). Further, this research investigated the consistency of model fit across time (Le., 1 month & 12 months) and samples, and explored the effectiveness of the ReachOut! Internet site as a psychoeducational prevention strategy for adolescent depression and suicidality. Gender, age, and location differences were also explored. Participants were 185 Year-9 students and 93 Year-10 students aged 14 - 16 years, from seven secondary schools in regional and rural Victoria. Students were given a survey which included the Parental Bonding Instrument (Parker, Tupling, & Brown, 1979), the Millon Adolescent Personality Inventory (Millon, Green, & Meagher, 1982), the Profile of Mood States Inventory (McNair & Lorr, 1964), items on suicidal behaviour including some questions from the Revised Adolescent Suicide Questionnaire (Pearce & Martin, 1994), and questions on loss and general demographics. Results supported an indirect model of risk factors, with family factors directly influencing personality factors, which in turn influenced mood factors, including depression, which then influenced suicidality. At the theoretical level, results supported Attachment Theory (Bowlby, 1969), demonstrating that perceived parenting styles that are warm and not overly controlling are more conducive to an adolescent's emotional well-being than are parenting styles that are cold and controlling. Further, results supported Millon's theory of personality (1981), demonstrating that parenting style influences a child's personality. Short-term intervention effects from the internet site were a decrease in Introversion for the full sample, and decreased Inhibition and Suicidality for a high-risk subgroup. Long-term age effects were decreased Inhibition and increased Anxiety for the fall sample. There was also a probable intervention effect for Depression for the high-risk subgroup. No location differences for the risk factors were found between regional and rural areas.

A bayesian nonparametric joint factor model for learning shared and individual subspaces from multiple data sources

Joint analysis of multiple data sources is becoming increasingly popular in transfer learning, multi-task learning and cross-domain data mining. One promising approach to model the data jointly is through learning the shared and individual factor subspaces. However, performance of this approach depends on the subspace dimensionalities and the level of sharing needs to be specified a priori. To this end, we propose a nonparametric joint factor analysis framework for modeling multiple related data sources. Our model utilizes the hierarchical beta process as a nonparametric prior to automatically infer the number of shared and individual factors. For posterior inference, we provide a Gibbs sampling scheme using auxiliary variables. The effectiveness of the proposed framework is validated through its application on two real world problems - transfer learning in text and image retrieval.