Gender bias in the Trends in Mathematics and Science Study 2003 (TIMMS) for Canadian students /

This study is a secondary data analysis of the Trends in Mathematics and Science Study 2003 (TIMSS) to determine if there is a gender bias, unbalanced number of items suited to the cognitive skill of one gender, and to compare performance by location. Results of the Grade 8, math portion of the test were examined. Items were coded as verbal, spatial, verbal /spatial or neither and as conventional or unconventional. A Kruskal- Wallis was completed for each category, comparing performance of students from Ontario, Quebec, and Singapore. A Factor Analysis was completed to determine if there were item categories with similar characteristics. Gender differences favouring males were found in the verbal conventional category for Canadian students and in the spatial conventional category for students in Quebec. The greatest differences were by location, as students in Singapore outperformed students from Canada in all areas except for the spatial unconventional category. Finally, whether an item is conventional or unconventional is more important than whether the item is verbal or spatial. Results show the importance of fair assessment for the genders in both the classroom and on standardized tests.

The effect of schooling on student self-concept /|nRonald S. Cowley. -- 260 St. Catharines [Ont. : s. n.],

The following research paper was a study into change in student academic and general self-concept with increase in grade level and age# The majority of literature found by this researcher dealt with self-concept and its relationship to achievement and interactions with others, Review, then, was in these two areas-particularily within the academic setting, but outside of it as well. It wrs hypothesized that there would be a decrease in both academic and general student self-concept with increase in grade level and age. Self-Appraisal inventories, measuring general and academic selfconcept, and Inferential Self-Reports, measuring only academic selfconcept, were the instruments used* Subjects were students, Trade 1 to 13, and ranging in age from 5 to 21„ Although al] Self-Appraisal inventories and all Self-1?eports were very similar, they differed according to three Grade levels: Primary (Grades 1 to 3)> Intermediate (Grades L to 8), and Secondary (Grades 9 to 13) • Students in the Primary division received only their respective Self-Appraisal inventory, while others v/ere administered both inventories designed for their grade level. Scores on the inventories were computed to percents and then mean percents were arrived at for epch grade, each of the three grade levels, each age, and each of three age intervals. In all of these instances Spearman1 s rank order coefficients (!lpff) were calculated and significance, at the *05 level, was determined by referring to a table of critical values for one-tailed tests* Similarily fftff scores were computed, but only for individual grades and ages, and significance was determined at the *0b level* In only one instance, the General Dimension for individual grades, was significance of overall decrease found* Consequently the hypotheses put forth did not gain support* The fltff scores, however, revealed some isolated significant changes for the Academic Dimension, which were generally decreases in mean percents from the last grade of one level to the first grade of the next* For age mean percents, significant changes generally took place at early (5 or 6) and late (20 or 21) ages* A number of reasons for the results were presented and were generally based upon the studentfs encounters, or lack of encounters, with achievement or success* No definite conclusions, relevant to the hypotheses stated^ could be made, although a number of isolated ones were drawn on the basis of significant fltff scores* As well, mention was made of the possible trends or tendencies that were revealed by the results, but that could not, or were not, proven significant by "t's11 or "p's"* Teaching methods stressing improvement in academic, as well as socially related, situations, were recommended and a model teaching approach was presented in Appendix B#

Verbal ability, previous practice and load on short-term memory as determiners of differences in a complex learning task: an experimental study

Traditional psychometric theory and practice classify people according to broad ability dimensions but do not examine how these mental processes occur. Hunt and Lansman (1975) proposed a 'distributed memory' model of cognitive processes with emphasis on how to describe individual differences based on the assumption that each individual possesses the same components. It is in the quality of these components ~hat individual differences arise. Carroll (1974) expands Hunt's model to include a production system (after Newell and Simon, 1973) and a response system. He developed a framework of factor analytic (FA) factors for : the purpose of describing how individual differences may arise from them. This scheme is to be used in the analysis of psychometric tes ts . Recent advances in the field of information processing are examined and include. 1) Hunt's development of differences between subjects designated as high or low verbal , 2) Miller's pursuit of the magic number seven, plus or minus two, 3) Ferguson's examination of transfer and abilities and, 4) Brown's discoveries concerning strategy teaching and retardates . In order to examine possible sources of individual differences arising from cognitive tasks, traditional psychometric tests were searched for a suitable perceptual task which could be varied slightly and administered to gauge learning effects produced by controlling independent variables. It also had to be suitable for analysis using Carroll's f ramework . The Coding Task (a symbol substitution test) found i n the Performance Scale of the WISe was chosen. Two experiments were devised to test the following hypotheses. 1) High verbals should be able to complete significantly more items on the Symbol Substitution Task than low verbals (Hunt, Lansman, 1975). 2) Having previous practice on a task, where strategies involved in the task may be identified, increases the amount of output on a similar task (Carroll, 1974). J) There should be a sUbstantial decrease in the amount of output as the load on STM is increased (Miller, 1956) . 4) Repeated measures should produce an increase in output over trials and where individual differences in previously acquired abilities are involved, these should differentiate individuals over trials (Ferguson, 1956). S) Teaching slow learners a rehearsal strategy would improve their learning such that their learning would resemble that of normals on the ,:same task. (Brown, 1974). In the first experiment 60 subjects were d.ivided·into high and low verbal, further divided randomly into a practice group and nonpractice group. Five subjects in each group were assigned randomly to work on a five, seven and nine digit code throughout the experiment. The practice group was given three trials of two minutes each on the practice code (designed to eliminate transfer effects due to symbol similarity) and then three trials of two minutes each on the actual SST task . The nonpractice group was given three trials of two minutes each on the same actual SST task . Results were analyzed using a four-way analysis of variance . In the second experiment 18 slow learners were divided randomly into two groups. one group receiving a planned strategy practioe, the other receiving random practice. Both groups worked on the actual code to be used later in the actual task. Within each group subjects were randomly assigned to work on a five, seven or nine digit code throughout. Both practice and actual tests consisted on three trials of two minutes each. Results were analyzed using a three-way analysis of variance . It was found in t he first experiment that 1) high or low verbal ability by itself did not produce significantly different results. However, when in interaction with the other independent variables, a difference in performance was noted . 2) The previous practice variable was significant over all segments of the experiment. Those who received previo.us practice were able to score significantly higher than those without it. J) Increasing the size of the load on STM severely restricts performance. 4) The effect of repeated trials proved to be beneficial. Generally, gains were made on each successive trial within each group. S) In the second experiment, slow learners who were allowed to practice randomly performed better on the actual task than subjeots who were taught the code by means of a planned strategy. Upon analysis using the Carroll scheme, individual differences were noted in the ability to develop strategies of storing, searching and retrieving items from STM, and in adopting necessary rehearsals for retention in STM. While these strategies may benef it some it was found that for others they may be harmful . Temporal aspects and perceptual speed were also found to be sources of variance within individuals . Generally it was found that the largest single factor i nfluencing learning on this task was the repeated measures . What e~ables gains to be made, varies with individuals . There are environmental factors, specific abilities, strategy development, previous learning, amount of load on STM , perceptual and temporal parameters which influence learning and these have serious implications for educational programs .

The influence of recreational ball hockey play on cardiovascular risk factors in sedentary males

Introduction: The prevalence of coronary artery disease (CAD) is ever increasing in western industrialized societies. An individuals overall risk for CAD may be quantified by integrating a number of factors including, but not limited to, cardiorespiratory fitness, body composition, blood lipid profile and blood pressure. It might be expected that interventions aimed at improving any or all of these independent factors might improve an individual 's overall risk. To this end, the influence of standard endurance type exercise on cardiorespiratory fitness, body composition, blood lipids and blood pressure, and by extension the reduction of coronary risk factors, is well documented. On the other hand, interval training (IT) has been shown to provide an extremely powerful stimulus for improving indices of cardiorespiratory function but the influence of this training type on coronary risk factors is unknown. Moreover, the vast majority of studies investigating the effects of IT on fitness have used laboratory type training protocols. As a result of this, the influence of participation in interval-type recreational sports on cardiorespiratory fitness and coronary risk factors is unknown. Aims: The aim of the present study was to evaluate the effectiveness of recreational ball hockey, a sport associated with interval-type activity patterns, on indices of aerobic function and coronary risk factors in sedentary men in the approximate age range of 30 - 60 years. Individual risk factors were compiled into an overall coronary risk factor score using the Framingham Point Scale (FPS). Methods: Twenty-four sedentary males (age range 30 - 60) participated in the study. Subject activity level was assessed apriori using questionnaire responses. All subjects (experimental and control) were assessed to have been inactive and sedentary prior to participation in the study. The experimental group (43 ± 3 years; 90 ± 3 kg) (n = 11) participated in one season of recreational ball hockey (our surrogate for IT). Member of this group played a total of 16 games during an 11 week span. During this time, the control group (43 ± 2 years; 89 ± 2 kg) (n = 11) performed no training and continued with their sedentary lifestyle. Prior to and following the ball hockey season, experimental and control subjects were tested for the following variables: 1) cardiorespiratory fitness (as V02 Max) 2) blood lipid profile 3) body composition 5) waist to hip ratio 6) blood glucose levels and 7) blood pressure. Subject V02 Max was assessed using the Rockport submaximal walking test on an indoor track. To assess body composition we determined body mass ratio (BMI), % body fat, % lean body mass and waist to hip ratio. The blood lipid profile included high density lipoprotein, low density lipoprotein and total cholesterol levels; in addition, the ratio of total cholesterol to high density was calculated. Blood triglycerides were also assessed. All data were analyzed using independent t - tests and all data are expressed as mean ± standard error. Statistical significance was accepted at p :S 0.05. Results: Pre-test values for all variables were similar between the experimental and control group. Moreover, although the intervention used in this study was associated with changes in some variables for subjects in the experimental group, subjects in the control group did not exhibit any changes over the same time period. BODY COMPOSITION: The % body fat of experimental subjects decreased by 4.6 ± 0.5%, from 28.1 ± 2.6 to 26.9 ± 2.5 % while that of the control group was unchanged at 22.7 ± 1.4 and 22.2 ± 1.3 %. However, lean body mass of experimental and control subjects did not change at 64.3 ± 1.3 versus 66.1 ± 1.3 kg and 65.5 ± 0.8 versus 64.7 ± 0.8 kg, respectively. In terms of body mass index and waist to hip ratio, neither the experimental nor the control group showed any significant change. Respective values for the waist to hip ratio and body mass index (pre and post) were as follows: 1 ± 0.1 vs 0.9 ± 0.1 (experimental) and 0.9 ± 0.1 versus 0.9 ± 0.1 (controls) while for BMI they were 29 ± 1.4 versus 29 ± 1.2 (experimental) and 26 ± 0.7 vs. 26 ± 0.7 (controls). CARDIORESPIRATORY FITNESS: In the experimental group, predicted values for absolute V02 Max increased by 10 ± 3% (i.e. 3.3 ± 0.1 to 3.6 ± 0.1 liters min -1 while that of control subjects did not change (3.4 ± 0.2 and 3.4 ± 0.2 liters min-I). In terms of relative values for V02 Max, the experimental group increased by 11 ± 2% (37 ± 1.4 to 41 ± 1.4 ml kg-l min-I) while that of control subjects did not change (41 ± 1.4 and 40 ± 1.4 ml kg-l min-I). BLOOD LIPIDS: Compared to pre-test values, post-test values for HDL were decreased by 14 ± 5 % in the experiment group (from 52.4 ± 4.4 to 45.2 ± 4.3 mg dl-l) while HDL data for the control group was unchanged (49.7 ± 3.6 and 48.3 ± 4.1 mg dl-l, respectively. On the other hand, LDL levels did not change for either the experimental or control group (110.2 ± 10.4 versus 112.3 ± 7.1 mg dl-1 and 106.1 ± 11.3 versus 127 ± 15.1 mg dl-1, respectively). Further, total cholesterol did not change in either the experimental or control group (181.3 ± 8.7 mg dl-1 versus 178.7± 4.9 mg dl-l) and 190.7 ± 12.2 versus 197.1 ± 16.1 mg dl-1, respectively). Similarly, the ratio of TC/HDL did not change for either the experimental or control group (3.8 ± 0.4 versus 4.5 ± 0.5 and 4 ± 0.4 versus 4.2 ± 0.4, respectively). Blood triglyceride levels were also not altered in either the experimental or control group (100.3 ± 19.6 versus 114.8 ± 15.3 mg dl-1 and 140 ± 23.5 versus 137.3 ± 17.9 mg dl-l, respectively). BLOOD GLUCOSE: Fasted blood glucose levels did not change in either the experimental or control group. Pre- and post-values for experimental and control groups were 92.5 ± 4.8 versus 93.3 ± 4.3 mg dl-l and 92.3 ± 11.3 versus 93.2 ± 2.6 mg dl-1 , respectively. BLOOD PRESSURE: No aspect of blood pressure was altered in either the experimental or control group. For example, pre- and post-test systolic blood pressures were 131 ± 2 versus 129 ± 2 mmHg (experimental) and 123 ± 2 and 125 ± 2 mmHg (controls), respectively. Pre- and post-test diastolic blood pressures were 84 ± 2 and 83 ± 2 mmHg (experimental) and 81 ± 1 versus 82 ± 1 mmHg, respectively. Similarly, calculated pulse pressure was not altered in the experimental or control as pre- and post-test values were 47 ± 1 versus 47 ± 2 mmlHg and 42 ± 2 versus 43 ± 2 mmHg, respectively. FRAMINGHAM POINT SCORE: The concerted changes reported above produced an increased risk in the Framingham Point Score for the subjects in the experimental group. For example, the pre- and post-test FPS increased from 1.4 ± 0.9 to 2.7 ± 0.7. On the other hand, pre- and post-test scores for the control group were 1.8 ± 1 versus 1.8 ± 0.9. Conclusions: Our data confirms previous studies showing that interval-type exercise is a useful intervention for increasing aerobic fitness. Moreover, the increase in V02 Max we found in response to limited participation in ball hockey (i.e. 16 games) suggests that recreational sport may help reduce this aspect of coronary risk in previously sedentary individual. On the other hand, our results showing little or no positive change in body composition, blood lipids or blood pressures suggest that one season of recreational sport in not in of itself a powerful enough stimulus to reduce the overall risk of coronary artery disease. In light of this, it is recommended that, in addition to participation in recreational sport, the performance of regular physical activity is used as an adjunct to provide a more powerful overall stimulus for decreasing coronary risk factors. LIMITATIONS: The increase in the FPS we found for the experimental group, indicative of an increased risk for coronary disease, was largely due to the large decrease in HDL we observed after compared to above one season of ball hockey. In light of the fact that cardiorespiratory fitness was increased and % body fat was decreased, as well as the fact that other parameters such as blood pressure showed positive (but non statistically significant) trends, the possibility that the decrease in HDL showed by our data was anomalous should be considered. FUTURE DIRECTIONS: The results of this study suggesting that recreational sport may be a potentially useful intervention in the reduction of CAD require to be corroborated by future studies specifically employing 1) more rigorous assessment of fitness and fitness change and 2) more prolonged or frequent participants.

Repertory grid adaptations in detecting connotative trends and social behaviour problems in education

Please consult the paper edition of this thesis to read. It is available on the 5th Floor of the Library at Call Number: Z 9999 E38 K66 1983

The Influence of Arousal on Moral Decision-making for Individuals with and without Mild Head Injury

Recent research has shown that University students with a history of self-reported mild head injury (MHI) are more willing to endorse moral transgressions associated with personal, relative to impersonal, dilemmas (Chiappetta & Good, 2008). However, the terms 'personal' and 'impersonal' in these dilemmas have functionally confounded the 'intentionality' of the transgression with the 'personal impact' or 'outcome' of the transgression. In this study we used a modified version of these moral dilemmas to investigate decision-making and sympathetic nervous system responsivity. Forty-eight University students (24 with MHI, 24 with no-MHI) read 24 scenarios depicting moral dilemmas varying as a function of 'intentionality' of the act (deliberate or unintentional) and its 'outcome' (physical harm, no physical harm, non-moral) and were required to rate their willingness to engage in the act. Physiological indices of arousal (e.g., heart rate - HR) were recorded throughout. Additionally, participants completed several neurocognitive tests. Results indicated significantly lowered HR activity at baseline, prior to, and during (but not after) making a decision for each type of dilemma for participants with MHI compared to their non-injured cohort. Further, they were more likely than their cohort to authorize personal injuries that were deliberately induced. MHI history was also associated with better performance on tasks of cognitive flexibility and attention; while students' complaints of postconcussive symptoms and their social problem solving abilities did not differ as a function of MHI history. The results provide subtle support for the hypothesis that both emotional and cognitive information guide moral decision making in ambiguous and emotionally distressing situations. Persons with even a MHI have diminished physiological arousal that may reflect disruption to the neural pathways of the VMPFC/OFC similar to those with more severe injuries.

A DYNAMIC FRAMEWORK FOR THE RELATIONS BETWEEN FOREIGN EXCHANGE RATES, SAVINGS AND INVESTMENTS

The Meese-Rogoff forecasting puzzle states that foreign exchange (FX) rates are unpredictable. Since one country’s macroeconomic conditions could affect the price of its national currency, we study the dynamic relations between the FX rates and some macroeconomic accounts. Our research tests whether the predictability of the FX rates could be improved through the advanced econometrics. Improving the predictability of the FX rates has important implications for various groups including investors, business entities and the government. The present thesis examines the dynamic relations between the FX rates, savings and investments for a sample of 25 countries from the Organization for Economic Cooperation and Development. We apply quarterly data of FX rates, macroeconomic indices and accounts including the savings and the investments over three decades. Through preliminary Augmented Dickey-Fuller unit root tests and Johansen cointegration tests, we found that the savings rate and the investment rate are cointegrated with the vector (1,-1). This result is consistent with many previous studies on the savings-investment relations and therefore confirms the validity of the Feldstein-Horioka puzzle. Because of the special cointegrating relation between the savings rate and investment rate, we introduce the savings-investment rate differential (SID). Investigating each country through a vector autoregression (VAR) model, we observe extremely insignificant coefficient estimates of the historical SIDs upon the present FX rates. We also report similar findings through the panel VAR approach. We thus conclude that the historical SIDs are useless in forecasting the FX rate. Nonetheless, the coefficients of the past FX rates upon the current SIDs for both the country-specific and the panel VAR models are statistically significant. Therefore, we conclude that the historical FX rates can conversely predict the SID to some degree. Specifically, depreciation in the domestic currency would cause the increase in the SID.

The Implementations of Professional Learning Communities (PLCs) in Beijing and Ontario Schools

Since the knowledge-based economy has become a fashion over the last few decades, the concept of the professional learning community (PLC) has started being accepted by educational institutions and governments as an effective framework to improve teachers’ collective work and collaboration. The purpose of this research was to compare and contrast the implementations of PLCs between Beijing schools and Ontario schools from principals’ personal narratives. In order to discover the lessons and widen the scope to understand the PLC, this research applied qualitative design to collect the data from two principal participants in each location by semistructured interviews. Four themes emerged: (a) structure and technology, (b) identity and climate, (c) task and support, and (d) change and challenge. This research found that the root of the characteristics of the PLCs in Beijing and Ontario was the different existing teaching and learning systems as well as the test systems. Teaching Research Groups (TRGs) is one of the systems that help Chinese to organize routine time and input resources to improve teachers’ professional development. However, Canadian schools lack a similar system that guarantees the time and resources. Moreover, standardized test plays different roles in China and Canada. In China, standardized tests, such as the college entrance examination, are regarded as the important purpose of education, whereas Ontario principals saw the Education Quality and Accountability Office (EQAO) as a tool rather than a primary purpose. These two main differences influenced principals’ beliefs, attitudes, strategies, and practices. The implications based on this discovery provide new perspectives for principals, teachers, policy makers, and scholars to widen and deepen the research and practice of the PLC.

The Association of Energy Intake with Body Mass in Children With and Without Probable Developmental Coordination Disorder

Objective To determine if there is an association between energy intake (EI) and overweight or obesity status (OWOB) in children with and without probable developmental coordination disorder (p-DCD). Methods 1905 children were included. The Bruininks-Oseretsky Test of Motor Proficiency was used to assess p-DCD, body mass index for OWOB, and the Harvard Food Frequency Questionnaire for EI. Comparative tests and logistic regressions were performed. Results Reported EI was similar between p-DCD and non-DCD children among boys (2291 vs. 2281 kcal/day, p=0.917), but much lower in p-DCD compared to non-DCD girls (1745 vs.. 2068 kcal/day, p=0.007). EI was negatively associated with OWOB in girls only (OR: 0.82 (0.68, 0.98)). Conclusions Girls with p-DCD have a lower reported EI compared to their non-DCD peers. EI is negatively associated with OWOB in girls with p-DCD. Future research is needed to assess longitudinally the potential impact of EI on OWOB in this population.

Potential Urinary miRNA Biomarker Candidates for the Accurate Detection of Prostate Cancer among Benign Prostatic Hyperplasia Patients

MicroRNAs (miRNAs) are a class of short (similar to 22nt), single stranded RNA molecules that function as post-transcriptional regulators of gene expression. MiRNAs can regulate a variety of important biological pathways, including: cellular proliferation, differentiation and apoptosis. Profiling of miRNA expression patterns was shown to be more useful than the equivalent mRNA profiles for characterizing poorly differentiated tumours. As such, miRNA expression "signatures" are expected to offer serious potential for diagnosing and prognosing cancers of any provenance. The aim of this study was to investigate the potential of using deregulation of urinary miRNAs in order to detect Prostate Cancer (PCa) among Benign Prostatic Hyperplasia (BPH). To identify the miRNA signatures specific for PCa, miRNA expression profiling of 8 PCa patients, 12 BPH patients and 10 healthy males was carried out using whole genome expression profiling. Differential expression of two individual miRNAs between healthy males and BPH patients was detected and found to possibly target genes related to PCa development and progression. The sensitivity and specificity of miR-1825 for detecting PCa among BPH individuals was found to be 60% and 69%, respectively. Whereas, the sensitivity and specificity of miR-484 were 80% and 19%, respectively. Additionally, the sensitivity and specificity for miR-1825/484 in tandem were 45% and 75%, respectively. The proposed PCa miRNA signatures may therefore be of great value for the accurate diagnosis of PCa and BPH. This exploratory study has identified several possible targets that merit further investigation towards the development and validation of diagnostically useful, non-invasive, urine-based tests that might not only help diagnose PCa but also possibly help differentiate it from BPH.

The Reluctance of Cattle to Change a Learned Choice May Confound Preference Tests

The conclusion of the article states "it appears that previously learned choices may affect future choices in Y-mazes for cattle. Another area that needs to be researched is the effects of a mildly aversive treatment versus a severely aversive treatment on the tendency of a bovine to resist changing a learned choice".

Simulation-Based Finite and Large Sample Tests in Multivariate Regressions.

In the context of multivariate linear regression (MLR) models, it is well known that commonly employed asymptotic test criteria are seriously biased towards overrejection. In this paper, we propose a general method for constructing exact tests of possibly nonlinear hypotheses on the coefficients of MLR systems. For the case of uniform linear hypotheses, we present exact distributional invariance results concerning several standard test criteria. These include Wilks' likelihood ratio (LR) criterion as well as trace and maximum root criteria. The normality assumption is not necessary for most of the results to hold. Implications for inference are two-fold. First, invariance to nuisance parameters entails that the technique of Monte Carlo tests can be applied on all these statistics to obtain exact tests of uniform linear hypotheses. Second, the invariance property of the latter statistic is exploited to derive general nuisance-parameter-free bounds on the distribution of the LR statistic for arbitrary hypotheses. Even though it may be difficult to compute these bounds analytically, they can easily be simulated, hence yielding exact bounds Monte Carlo tests. Illustrative simulation experiments show that the bounds are sufficiently tight to provide conclusive results with a high probability. Our findings illustrate the value of the bounds as a tool to be used in conjunction with more traditional simulation-based test methods (e.g., the parametric bootstrap) which may be applied when the bounds are not conclusive.

Exact Tests for Contemporaneous Correlation of Disturbances in Seemingly Unrelated Regressions.

This paper proposes finite-sample procedures for testing the SURE specification in multi-equation regression models, i.e. whether the disturbances in different equations are contemporaneously uncorrelated or not. We apply the technique of Monte Carlo (MC) tests [Dwass (1957), Barnard (1963)] to obtain exact tests based on standard LR and LM zero correlation tests. We also suggest a MC quasi-LR (QLR) test based on feasible generalized least squares (FGLS). We show that the latter statistics are pivotal under the null, which provides the justification for applying MC tests. Furthermore, we extend the exact independence test proposed by Harvey and Phillips (1982) to the multi-equation framework. Specifically, we introduce several induced tests based on a set of simultaneous Harvey/Phillips-type tests and suggest a simulation-based solution to the associated combination problem. The properties of the proposed tests are studied in a Monte Carlo experiment which shows that standard asymptotic tests exhibit important size distortions, while MC tests achieve complete size control and display good power. Moreover, MC-QLR tests performed best in terms of power, a result of interest from the point of view of simulation-based tests. The power of the MC induced tests improves appreciably in comparison to standard Bonferroni tests and, in certain cases, outperforms the likelihood-based MC tests. The tests are applied to data used by Fischer (1993) to analyze the macroeconomic determinants of growth.

Économétrie, théorie des tests et philosophie des sciences

Dans ce texte, nous revoyons certains développements récents de l’économétrie qui peuvent être intéressants pour des chercheurs dans des domaines autres que l’économie et nous soulignons l’éclairage particulier que l’économétrie peut jeter sur certains thèmes généraux de méthodologie et de philosophie des sciences, tels la falsifiabilité comme critère du caractère scientifique d’une théorie (Popper), la sous-détermination des théories par les données (Quine) et l’instrumentalisme. En particulier, nous soulignons le contraste entre deux styles de modélisation - l’approche parcimonieuse et l’approche statistico-descriptive - et nous discutons les liens entre la théorie des tests statistiques et la philosophie des sciences.

Simulation-Based Finite-Sample Tests for Heteroskedasticity and ARCH Effects.

A wide range of tests for heteroskedasticity have been proposed in the econometric and statistics literature. Although a few exact homoskedasticity tests are available, the commonly employed procedures are quite generally based on asymptotic approximations which may not provide good size control in finite samples. There has been a number of recent studies that seek to improve the reliability of common heteroskedasticity tests using Edgeworth, Bartlett, jackknife and bootstrap methods. Yet the latter remain approximate. In this paper, we describe a solution to the problem of controlling the size of homoskedasticity tests in linear regression contexts. We study procedures based on the standard test statistics [e.g., the Goldfeld-Quandt, Glejser, Bartlett, Cochran, Hartley, Breusch-Pagan-Godfrey, White and Szroeter criteria] as well as tests for autoregressive conditional heteroskedasticity (ARCH-type models). We also suggest several extensions of the existing procedures (sup-type of combined test statistics) to allow for unknown breakpoints in the error variance. We exploit the technique of Monte Carlo tests to obtain provably exact p-values, for both the standard and the new tests suggested. We show that the MC test procedure conveniently solves the intractable null distribution problem, in particular those raised by the sup-type and combined test statistics as well as (when relevant) unidentified nuisance parameter problems under the null hypothesis. The method proposed works in exactly the same way with both Gaussian and non-Gaussian disturbance distributions [such as heavy-tailed or stable distributions]. The performance of the procedures is examined by simulation. The Monte Carlo experiments conducted focus on : (1) ARCH, GARCH, and ARCH-in-mean alternatives; (2) the case where the variance increases monotonically with : (i) one exogenous variable, and (ii) the mean of the dependent variable; (3) grouped heteroskedasticity; (4) breaks in variance at unknown points. We find that the proposed tests achieve perfect size control and have good power.