Multicollinearity and financial constraint in investment decisions: a Bayesian generalized ridge regression

This paper addresses the investment decisions considering the presence of financial constraints of 373 large Brazilian firms from 1997 to 2004, using panel data. A Bayesian econometric model was used considering ridge regression for multicollinearity problems among the variables in the model. Prior distributions are assumed for the parameters, classifying the model into random or fixed effects. We used a Bayesian approach to estimate the parameters, considering normal and Student t distributions for the error and assumed that the initial values for the lagged dependent variable are not fixed, but generated by a random process. The recursive predictive density criterion was used for model comparisons. Twenty models were tested and the results indicated that multicollinearity does influence the value of the estimated parameters. Controlling for capital intensity, financial constraints are found to be more important for capital-intensive firms, probably due to their lower profitability indexes, higher fixed costs and higher degree of property diversification.

Bayesian analysis of the error correction model

This paper presents a method for estimating the posterior probability density of the cointegrating rank of a multivariate error correction model. A second contribution is the careful elicitation of the prior for the cointegrating vectors derived from a prior on the cointegrating space. This prior obtains naturally from treating the cointegrating space as the parameter of interest in inference and overcomes problems previously encountered in Bayesian cointegration analysis. Using this new prior and Laplace approximation, an estimator for the posterior probability of the rank is given. The approach performs well compared with information criteria in Monte Carlo experiments. (C) 2003 Elsevier B.V. All rights reserved.

FibroTest is an independent predictor of virologic response in chronic hepatitis C patients retreated with pegylated interferon alfa-2b and ribavirin in the EPIC(3) program

Background & Aims: EPIC-3 is a prospective, international study that has demonstrated the efficacy of PEG-IFN alfa-2b plus weight-based ribavirin in patients with chronic hepatitis C and significant fibrosis who previously failed any interferon-alfa/ribavirin therapy. The aim of the present study was to assess FibroTest (FT), a validated non-invasive marker of fibrosis in treatment-naive patients, as a possible alternative to biopsy as the baseline predictor of subsequent early virologic (EVR) and sustained virologic response (SVR) in previously treated patients. Methods: Of 2312 patients enrolled, 1459 had an available baseline FT, biopsy, and complete data. Uni- (UV) and multi-variable (MV) analyses were performed using FT and biopsy. Results: Baseline characteristics were similar as in the overall population; METAVIR stage: 28% F2, 29% F3, and 43% F4, previous relapsers 29%, previous PEG-IFN regimen 41%, high baseline viral load (BVL) 64%. 506 patients (35%) had undetectable HCV-RNA at TW12 (TW12neg), with 58% achieving SVR. The accuracy of FT was similar to that in naive patients: AUROC curve for the diagnosis of F4 vs F2 = 0.80 (p<0.00001). Five baseline factors were associated (p<0.001) with SVR in UV and MV analyses (odds ratio: UV/MV): fibrosis stage estimated using FT (4.5/5.9) or biopsy (1.5/1.6), genotype 2/3 (4.5/5.1), BVL (1.5/1.3), prior relapse (1.6/1.6), previous treatment with non-PEG-IFN (2.6/2.0). These same factors were associated (p <= 0.001) with EVR. Among patients TW12neg, two independent factors remained highly predictive of SVR by MV analysis (p <= 0.001): genotype 2/3 (odds ratio = 2.9), fibrosis estimated with FT (4.3) or by biopsy (1.5). Conclusions: FibroTest at baseline is a possible non-invasive alternative to biopsy for the prediction of EVR at 12 weeks and SVR, in patients with previous failures and advanced fibrosis, retreated with PEG-IFN alfa-2b and ribavirin. (C) 2010 European Association for the Study of the Liver. Published by Elsevier B.V. All rights reserved.

Molecular characterization of the Hepatitis B virus genotypes in Colombia: A Bayesian inference on the genotype F

Hepatitis B is a worldwide health problem affecting about 2 billion people and more than 350 million are chronic carriers of the virus. Nine HBV genotypes (A to I) have been described. The geographical distribution of HBV genotypes is not completely understood due to the limited number of samples from some parts of the world. One such example is Colombia, in which few studies have described the HBV genotypes. In this study, we characterized HBV genotypes in 143 HBsAg-positive volunteer blood donors from Colombia. A fragment of 1306 bp partially comprising HBsAg and the DNA polymerase coding regions (S/POL) was amplified and sequenced. Bayesian phylogenetic analyses were conducted using the Markov Chain Monte Carlo (MCMC) approach to obtain the maximum clade credibility (MCC) tree using BEAST v.1.5.3. Of all samples, 68 were positive and 52 were successfully sequenced. Genotype F was the most prevalent in this population (77%) - subgenotypes F3 (75%) and Fib (2%). Genotype G (7.7%) and subgenotype A2 (15.3%) were also found. Genotype G sequence analysis suggests distinct introductions of this genotype in the country. Furthermore, we estimated the time of the most recent common ancestor (TMRCA) for each HBV/F subgenotype and also for Colombian F3 sequences using two different datasets: (i) 77 sequences comprising 1306 bp of S/POL region and (ii) 283 sequences comprising 681 bp of S/POL region. We also used two other previously estimated evolutionary rates: (i) 2.60 x 10(-4) s/s/y and (ii) 1.5 x 10(-5) s/s/y. Here we report the HBV genotypes circulating in Colombia and estimated the TMRCA for the four different subgenotypes of genotype F. (C) 2010 Elsevier B.V. All rights reserved.

Peginterferon alfa-2b and Ribavirin: Effective in Patients With Hepatitis C Who Failed Interferon alfa/Ribavirin Therapy

Background & Aims: Treatment with peginterferon alfa and ribavirin produces a sustained virologic response (SVR) in approximately 60% of hepatitis C virus (HCV)-infected patients. Alternate options are needed for patients who relapse or do not respond to therapy. Methods: This prospective, international, multicenter, open-label study evaluated efficacy and safety of peginterferon alfa-2b (1.5 mu g/kg/wk) plus weight-based ribavirin (800-1400 mg/day) in 2333 chronic HCV-infected patients with significant fibrosis/cirrhosis whose previous interferon alfa/ribavirin therapy failed. Patients with undetectable HCV-RNA at treatment week (TW) 12 received 48 weeks of therapy; patients with detectable HCV-RNA at TW12 could enter maintenance studies at TW18; 188 patients with low/detectable HCV-RNA at TW12 continued therapy at the investigator`s request. Results: Overall, 22% of the patients attained SVR (56% with undetectable HCV-RNA and 12% with low/detectable HCV-RNA at TW12). SVR was better in relapsers (38%) than nonresponders (14%), regardless of previous treatment, and in patients previously treated with interferon-alfa/ribavirin (25%) than peginterferon alfa-ribavirin (17%). Predictors of response in patients with undetectable HCV-RNA at TW12 were genotype (2/3 vs 1, respectively; odds ratio [OR] 2.4; P < .0001), fibrosis score (F2 vs F4; OR, 2.2; F3 vs F4; OR, 1.7; P < .0001), and baseline viral load (<= 600,000 vs >600,000 IU/mL; OR, 1.4; P = .0223). These factors plus previous treatment and response were overall predictors of SVR. Safety was similar among fibrosis groups. Conclusions: Peginterferon alfa-2b plus weight-based ribavirin is effective and safe in patients who failed interferon alfa/ribavirin therapy. Genotype, baseline viral load, and fibrosis stage were predictors of response.

Very early prediction of response to HCV treatment with PEG-IFN-alfa-2a and ribavirin in HIV/HCV-coinfected patients

The objective of this study was to find very early viral kinetic markers to predict nonresponse to hepatitis C virus (HCV) therapy in a group of human immunodeficiency virus (HIV)/HCV-coinfected patients. Twenty-six patients (15 HCV genotype-1 and 11 genotype-3) were treated with a 48-week regimen of peginterferon-alfa-2a (PEG-IFN) (180 mu g/week) and weight-based ribavirin (11 mg/kg/day). Samples were collected at baseline; 4, 8, 12, 18, 24, 30, 36 and 42 h; days 2, 3, 4, 7, 8, 15, 22, 29, 43 and 57 then weekly and monthly. Five patients discontinued treatment. Seven patients (27%) achieved a sustained virological response (SVR). Nadir HCV RNA levels were observed 1.6 +/- 0.3 days after initiation of therapy, followed by a 0.3- to 12.9-fold viral rebound until the administration of the second dose of PEG-IFN, which were not associated with SVR or HCV genotype. A viral decline < 1.19 log for genotype-1 and < 0.97 log for genotype-3, 2 days after starting therapy, had a negative predictive value (NPV) of 100% for SVR. The day 2 virological response had a similar positive predictive value for SVR as a rapid virological response at week 4. In addition, a second-phase viral decline slope (i.e., measured from day 2 to 29) < 0.3 log/week had a NPV = 100% for SVR. We conclude that first-phase viral decline at day 2 and second-phase viral decline slope (< 0.3 log/week) are excellent predictors of nonresponse. Further studies are needed to validate these viral kinetic parameters as early on-treatment prognosticators of nonresponse in patients with HCV and HIV.

A Bayesian approach to fuzzy hypotheses testing for the estimation of optimal age for vaccination against measles

Fuzzy Bayesian tests were performed to evaluate whether the mother`s seroprevalence and children`s seroconversion to measles vaccine could be considered as ""high"" or ""low"". The results of the tests were aggregated into a fuzzy rule-based model structure, which would allow an expert to influence the model results. The linguistic model was developed considering four input variables. As the model output, we obtain the recommended age-specific vaccine coverage. The inputs of the fuzzy rules are fuzzy sets and the outputs are constant functions, performing the simplest Takagi-Sugeno-Kang model. This fuzzy approach is compared to a classical one, where the classical Bayes test was performed. Although the fuzzy and classical performances were similar, the fuzzy approach was more detailed and revealed important differences. In addition to taking into account subjective information in the form of fuzzy hypotheses it can be intuitively grasped by the decision maker. Finally, we show that the Bayesian test of fuzzy hypotheses is an interesting approach from the theoretical point of view, in the sense that it combines two complementary areas of investigation, normally seen as competitive. (C) 2007 IMACS. Published by Elsevier B.V. All rights reserved.

A Sustained Virologic Response Is Durable in Patients With Chronic Hepatitis C Treated With Peginterferon Alfa-2a and Ribavirin

BACKGROUND & AIMS: A sustained virologic response (SVR) to therapy for hepatitis C virus (HCV) infection is defined as the inability to detect HCV RNA 24 weeks after completion of treatment. Although small studies have reported that the SVR is durable and lasts for long periods, it has not been conclusively shown. METHODS: The durability of treatment responses was examined in patients originally enrolled in one of 9 randomized multicenter trials (n = 1343). The study included patients who received pegylated interferon (peginterferon) alfa-2a alone (n = 166) or in combination with ribavirin (n = 1077, including 79 patients with normal alanine aminotransferase levels and 100 patients who were coinfected with human immunodeficiency virus and HCV) and whose serum samples were negative for HCV RNA (<50 IU/mL) at their final assessment. Patients were assessed annually, from the date of last treatment, for a mean of 3.9 years (range, 0.8-7.1 years). RESULTS: Most patients (99.1%) who achieved an SVR had undetectable levels of HCV RNA in serum samples throughout the follow-up period. Serum samples from 0.9% of the patients contained HCV RNA a mean of 1.8 years (range, 1.1-2.9 years) after treatment ended. It is not clear if these patients were reinfected or experienced a relapse. CONCLUSIONS: In a large cohort of patients monitored for the durability of an SVR, the SVR was maintained for almost 4 years after treatment with peginterferon alfa-2a alone or in combination with ribavirin. In patients with chronic hepatitis C infection, the SVR is durable and these patients should be considered as cured.

Peginterferon alfa-2a plus ribavirin for chronic hepatitis C virus infection in children and adolescents

Background & Aims: This multi-center study aimed to prospectively evaluate the safety and efficacy of a genotype-based Pegylated Interferon alfa-2a/Ribavirin therapy in treatment-naive hepatitis C virus (HCV), positive HCV serology, and quantifiable HCV RNA, infected children. Methods: Eighteen children with genotypes 2 and 3 patients (group A) were assigned to medication for 24 weeks, and 47 children with genotypes 1, 4, 5 and 6 patients (group B) for 48 weeks. Results: Early response at week 12 was observed in 83% of group A patients and in 57% of group B patients (p <0.05). End of treatment response was achieved in 94% of patients in group A and in 57% in group B (p <0.001). Sustained virologic response was maintained in 89% of patients in group A and in 57% of patients in group B (p <0.01). Ten patients stopped prematurely the treatment, 2 for serious adverse event (acute hepatitis and thyrotoxicosis), and 8 because of no virologic response at week 24. Peginterferon alfa-2a and Ribavirin dose was adjusted in 15 patients (23%), 11 for neutropenia (17%), and 3 patients (5%), for anemia, respectively. Treatment-related adverse events included fever and flu-like symptoms (54%), irritability depression change of mood (34%), vomiting (23%), abdominal pain (38%), loss of appetite (21.5%) and dermatitis (29%). No influence on height growth was observed. Conclusions: Pegylated inteferon alfa-2a and Ribavirin treatment allowed to achieve SVR in 57% of pediatric patients with genotypes 1, 4, 5 and 6, and in 94% of genotypes 2 and 3. These results show an improved SVR as compared to reference series in adults with similar regimen. (c) 2010 European Association for the Study of the Liver. Published by Elsevier B.V. All rights reserved.

Kidney Function and 24-Hour Proteinuria in Patients with Fabry Disease during 36 Months of Agalsidase Alfa Enzyme Replacement Therapy: A Brazilian Experience

Background. Prior to the introduction of enzyme replacement therapy (ERT), management of Fabry disease (FD) consisted of symptomatic and palliative measures. ERT has been available for several years using recombinant human agalsidase alfa, an analogue of alpha-galactosidase A (GALA). However, the limitations of ERT in improving kidney function have not been established. This study evaluates the safety and therapeutic effect of agalsidase alfa replacement in terms of kidney function and reduction in 24-hour proteinuria. Methods. During the period between January 1, 2002, and August 1, 2005, nine Fabry patients (7 male, 2 female) were treated according to protocol, receiving 0.2 mg/kg agalsidase alfa IV every two weeks. Kidney function was evaluated by measuring the glomerular filtration rate (GFR) using chromium ethylene diamine tetra-acetate clearance ((51)Cr-EDTA mL/min/1.73 m(2)) at baseline, 12, 24, and 36 months. 24-hour proteinuria was measured at baseline, 3, 6, 12, 18, 24, and 36 months of ERT. Kidney disease was classified according to National Kidney Foundation Disease Outcome Quality Initiative (NKF/DOQI) Advisory Board criteria, which define stage I chronic kidney disease (CKD) as GFR >= 90mL/min/1.73 m(2), stage II as 60-89 mL/min/1.73m(2), stage III as 30-59 mL/min/1.73 m(2), stage IV as 15-29 mL/min/1.73m(2), and stage V as < 15 mL/min/1.73m(2). Results. Six patients completed 36 months of therapy, 2 patients completed 18 months, and 1 patient completed 12 months. Mean patient age at baseline was 34.6 +/- 11.3 years. During the study period, kidney function remained stable in patients with stages I, II, or III CKD. One patient, who entered the study with stage IV CKD, progressed to end-stage chronic kidney disease, beginning hemodialysis after 7 months and receiving a kidney transplant after 12 months of ERT. Proteinuria also remained stable in the group of patients with pathologic proteinuria. The use of agalsidase alfa was well tolerated in 99.5% of the infusions administered. Conclusion. Over the course of 36 months of ERT, there was no change in kidney function and 24-hour proteinuria. This suggests thatagalsidase alfa may slow or halt the progression of kidney disease when used before extensive kidney damage occurs. No significant side effects were observed with ERT during the course of the study.

A Comparison of Bayesian and Sampling Theory Inferences in a Probit Model

HE PROBIT MODEL IS A POPULAR DEVICE for explaining binary choice decisions in econometrics. It has been used to describe choices such as labor force participation, travel mode, home ownership, and type of education. These and many more examples can be found in papers by Amemiya (1981) and Maddala (1983). Given the contribution of economics towards explaining such choices, and given the nature of data that are collected, prior information on the relationship between a choice probability and several explanatory variables frequently exists. Bayesian inference is a convenient vehicle for including such prior information. Given the increasing popularity of Bayesian inference it is useful to ask whether inferences from a probit model are sensitive to a choice between Bayesian and sampling theory techniques. Of interest is the sensitivity of inference on coefficients, probabilities, and elasticities. We consider these issues in a model designed to explain choice between fixed and variable interest rate mortgages. Two Bayesian priors are employed: a uniform prior on the coefficients, designed to be noninformative for the coefficients, and an inequality restricted prior on the signs of the coefficients. We often know, a priori, whether increasing the value of a particular explanatory variable will have a positive or negative effect on a choice probability. This knowledge can be captured by using a prior probability density function (pdf) that is truncated to be positive or negative. Thus, three sets of results are compared:those from maximum likelihood (ML) estimation, those from Bayesian estimation with an unrestricted uniform prior on the coefficients, and those from Bayesian estimation with a uniform prior truncated to accommodate inequality restrictions on the coefficients.