Bayesian forecasting with highly correlated predictors

This paper considers Bayesian variable selection in regressions with a large number of possibly highly correlated macroeconomic predictors. I show that by acknowledging the correlation structure in the predictors can improve forecasts over existing popular Bayesian variable selection algorithms.

Improvement of therapeutic drug monitoring of imatinib by Bayesian prediction of trough levels. Personalized drug dosage

Aims: Plasma concentrations of imatinib differ largely between patients despite same dosage, owing to large inter-individual variability in pharmacokinetic (PK) parameters. As the drug concentration at the end of the dosage interval (Cmin) correlates with treatment response and tolerability, monitoring of Cmin is suggested for therapeutic drug monitoring (TDM) of imatinib. Due to logistic difficulties, random sampling during the dosage interval is however often performed in clinical practice, thus rendering the respective results not informative regarding Cmin values.Objectives: (I) To extrapolate randomly measured imatinib concentrations to more informative Cmin using classical Bayesian forecasting. (II) To extend the classical Bayesian method to account for correlation between PK parameters. (III) To evaluate the predictive performance of both methods.Methods: 31 paired blood samples (random and trough levels) were obtained from 19 cancer patients under imatinib. Two Bayesian maximum a posteriori (MAP) methods were implemented: (A) a classical method ignoring correlation between PK parameters, and (B) an extended one accounting for correlation. Both methods were applied to estimate individual PK parameters, conditional on random observations and covariate-adjusted priors from a population PK model. The PK parameter estimates were used to calculate trough levels. Relative prediction errors (PE) were analyzed to evaluate accuracy (one-sample t-test) and to compare precision between the methods (F-test to compare variances).Results: Both Bayesian MAP methods allowed non-biased predictions of individual Cmin compared to observations: (A) - 7% mean PE (CI95% - 18 to 4 %, p = 0.15) and (B) - 4% mean PE (CI95% - 18 to 10 %, p = 0.69). Relative standard deviations of actual observations from predictions were 22% (A) and 30% (B), i.e. comparable to the intraindividual variability reported. Precision was not improved by taking into account correlation between PK parameters (p = 0.22).Conclusion: Clinical interpretation of randomly measured imatinib concentrations can be assisted by Bayesian extrapolation to maximum likelihood Cmin. Classical Bayesian estimation can be applied for TDM without the need to include correlation between PK parameters. Both methods could be adapted in the future to evaluate other individual pharmacokinetic measures correlated to clinical outcomes, such as area under the curve(AUC).

Implementation of bayesian therapeutic drug monitoring in modern patient care

The variability observed in drug exposure has a direct impact on the overall response to drug. The largest part of variability between dose and drug response resides in the pharmacokinetic phase, i.e. in the dose-concentration relationship. Among possibilities offered to clinicians, Therapeutic Drug Monitoring (TDM; Monitoring of drug concentration measurements) is one of the useful tool to guide pharmacotherapy. TDM aims at optimizing treatments by individualizing dosage regimens based on blood drug concentration measurement. Bayesian calculations, relying on population pharmacokinetic approach, currently represent the gold standard TDM strategy. However, it requires expertise and computational assistance, thus limiting its large implementation in routine patient care. The overall objective of this thesis was to implement robust tools to provide Bayesian TDM to clinician in modern routine patient care. To that endeavour, aims were (i) to elaborate an efficient and ergonomic computer tool for Bayesian TDM: EzeCHieL (ii) to provide algorithms for drug concentration Bayesian forecasting and software validation, relying on population pharmacokinetics (iii) to address some relevant issues encountered in clinical practice with a focus on neonates and drug adherence. First, the current stage of the existing software was reviewed and allows establishing specifications for the development of EzeCHieL. Then, in close collaboration with software engineers a fully integrated software, EzeCHieL, has been elaborated. EzeCHieL provides population-based predictions and Bayesian forecasting and an easy-to-use interface. It enables to assess the expectedness of an observed concentration in a patient compared to the whole population (via percentiles), to assess the suitability of the predicted concentration relative to the targeted concentration and to provide dosing adjustment. It allows thus a priori and a posteriori Bayesian drug dosing individualization. Implementation of Bayesian methods requires drug disposition characterisation and variability quantification trough population approach. Population pharmacokinetic analyses have been performed and Bayesian estimators have been provided for candidate drugs in population of interest: anti-infectious drugs administered to neonates (gentamicin and imipenem). Developed models were implemented in EzeCHieL and also served as validation tool in comparing EzeCHieL concentration predictions against predictions from the reference software (NONMEM®). Models used need to be adequate and reliable. For instance, extrapolation is not possible from adults or children to neonates. Therefore, this work proposes models for neonates based on the developmental pharmacokinetics concept. Patients' adherence is also an important concern for drug models development and for a successful outcome of the pharmacotherapy. A last study attempts to assess impact of routine patient adherence measurement on models definition and TDM interpretation. In conclusion, our results offer solutions to assist clinicians in interpreting blood drug concentrations and to improve the appropriateness of drug dosing in routine clinical practice.

Três ensaios sobre política monetária e crédito

In the first essay, "Determinants of Credit Expansion in Brazil", analyzes the determinants of credit using an extensive bank level panel dataset. Brazilian economy has experienced a major boost in leverage in the first decade of 2000 as a result of a set factors ranging from macroeconomic stability to the abundant liquidity in international financial markets before 2008 and a set of deliberate decisions taken by President Lula's to expand credit, boost consumption and gain political support from the lower social strata. As relevant conclusions to our investigation we verify that: credit expansion relied on the reduction of the monetary policy rate, international financial markets are an important source of funds, payroll-guaranteed credit and investment grade status affected positively credit supply. We were not able to confirm the importance of financial inclusion efforts. The importance of financial sector sanity indicators of credit conditions cannot be underestimated. These results raise questions over the sustainability of this expansion process and financial stability in the future. The second essay, “Public Credit, Monetary Policy and Financial Stability”, discusses the role of public credit. The supply of public credit in Brazil has successfully served to relaunch the economy after the Lehman-Brothers demise. It was later transformed into a driver for economic growth as well as a regulation device to force private banks to reduce interest rates. We argue that the use of public funds to finance economic growth has three important drawbacks: it generates inflation, induces higher loan rates and may induce financial instability. An additional effect is the prevention of market credit solutions. This study contributes to the understanding of the costs and benefits of credit as a fiscal policy tool. The third essay, “Bayesian Forecasting of Interest Rates: Do Priors Matter?”, discusses the choice of priors when forecasting short-term interest rates. Central Banks that commit to an Inflation Target monetary regime are bound to respond to inflation expectation spikes and product hiatus widening in a clear and transparent way by abiding to a Taylor rule. There are various reports of central banks being more responsive to inflationary than to deflationary shocks rendering the monetary policy response to be indeed non-linear. Besides that there is no guarantee that coefficients remain stable during time. Central Banks may switch to a dual target regime to consider deviations from inflation and the output gap. The estimation of a Taylor rule may therefore have to consider a non-linear model with time varying parameters. This paper uses Bayesian forecasting methods to predict short-term interest rates. We take two different approaches: from a theoretic perspective we focus on an augmented version of the Taylor rule and include the Real Exchange Rate, the Credit-to-GDP and the Net Public Debt-to-GDP ratios. We also take an ”atheoretic” approach based on the Expectations Theory of the Term Structure to model short-term interest. The selection of priors is particularly relevant for predictive accuracy yet, ideally, forecasting models should require as little a priori expert insight as possible. We present recent developments in prior selection, in particular we propose the use of hierarchical hyper-g priors for better forecasting in a framework that can be easily extended to other key macroeconomic indicators.

Toward better outcomes with tacrolimus therapy: Population pharmacokinetics and individualized dosage prediction in adult liver transplantation

Patient outcomes in transplantation would improve if dosing of immunosuppressive agents was individualized. The aim of this study is to develop a population pharmacokinetic model of tacrolimus in adult liver transplant recipients and test this model in individualizing therapy. Population analysis was performed on data from 68 patients. Estimates were sought for apparent clearance (CL/F) and apparent volume of distribution (V/F) using the nonlinear mixed effects model program (NONMEM). Factors screened for influence on these parameters were weight, age, sex, transplant type, biliary reconstructive procedure, postoperative day, days of therapy, liver function test results, creatinine clearance, hematocrit, corticosteroid dose, and interacting drugs. The predictive performance of the developed model was evaluated through Bayesian forecasting in an independent cohort of 36 patients. No linear correlation existed between tacrolimus dosage and trough concentration (r(2) = 0.005). Mean individual Bayesian estimates for CL/F and V/F were 26.5 8.2 (SD) L/hr and 399 +/- 185 L, respectively. CL/F was greater in patients with normal liver function. V/F increased with patient weight. CL/F decreased with increasing hematocrit. Based on the derived model, a 70-kg patient with an aspartate aminotransferase (AST) level less than 70 U/L would require a tacrolimus dose of 4.7 mg twice daily to achieve a steady-state trough concentration of 10 ng/mL. A 50-kg patient with an AST level greater than 70 U/L would require a dose of 2.6 mg. Marked interindividual variability (43% to 93%) and residual random error (3.3 ng/mL) were observed. Predictions made using the final model were reasonably nonbiased (0.56 ng/mL), but imprecise (4.8 ng/mL). Pharmacokinetic information obtained will assist in tacrolimus dosing; however, further investigation into reasons for the pharmacokinetic variability of tacrolimus is required.

Clinical pharmacokinetics and pharmacodynamics of tacrolimus in solid organ transplantation

The aim of this review is to analyse critically the recent literature on the clinical pharmacokinetics and pharmacodynamics of tacrolimus in solid organ transplant recipients. Dosage and target concentration recommendations for tacrolimus vary from centre to centre, and large pharmacokinetic variability makes it difficult to predict what concentration will be achieved with a particular dose or dosage change. Therapeutic ranges have not been based on statistical approaches. The majority of pharmacokinetic studies have involved intense blood sampling in small homogeneous groups in the immediate post-transplant period. Most have used nonspecific immunoassays and provide little information on pharmacokinetic variability. Demographic investigations seeking correlations between pharmacokinetic parameters and patient factors have generally looked at one covariate at a time and have involved small patient numbers. Factors reported to influence the pharmacokinetics of tacrolimus include the patient group studied, hepatic dysfunction, hepatitis C status, time after transplantation, patient age, donor liver characteristics, recipient race, haematocrit and albumin concentrations, diurnal rhythm, food administration, corticosteroid dosage, diarrhoea and cytochrome P450 (CYP) isoenzyme and P-glycoprotein expression. Population analyses are adding to our understanding of the pharmacokinetics of tacrolimus, but such investigations are still in their infancy. A significant proportion of model variability remains unexplained. Population modelling and Bayesian forecasting may be improved if CYP isoenzymes and/or P-glycoprotein expression could be considered as covariates. Reports have been conflicting as to whether low tacrolimus trough concentrations are related to rejection. Several studies have demonstrated a correlation between high trough concentrations and toxicity, particularly nephrotoxicity. The best predictor of pharmacological effect may be drug concentrations in the transplanted organ itself. Researchers have started to question current reliance on trough measurement during therapeutic drug monitoring, with instances of toxicity and rejection occurring when trough concentrations are within 'acceptable' ranges. The correlation between blood concentration and drug exposure can be improved by use of non-trough timepoints. However, controversy exists as to whether this will provide any great benefit, given the added complexity in monitoring. Investigators are now attempting to quantify the pharmacological effects of tacrolimus on immune cells through assays that measure in vivo calcineurin inhibition and markers of immuno suppression such as cytokine concentration. To date, no studies have correlated pharmacodynamic marker assay results with immunosuppressive efficacy, as determined by allograft outcome, or investigated the relationship between calcineurin inhibition and drug adverse effects. Little is known about the magnitude of the pharmacodynamic variability of tacrolimus.

Forecasting with Medium and Large Bayesian VARs

This paper is motivated by the recent interest in the use of Bayesian VARs for forecasting, even in cases where the number of dependent variables is large. In such cases, factor methods have been traditionally used but recent work using a particular prior suggests that Bayesian VAR methods can forecast better. In this paper, we consider a range of alternative priors which have been used with small VARs, discuss the issues which arise when they are used with medium and large VARs and examine their forecast performance using a US macroeconomic data set containing 168 variables. We nd that Bayesian VARs do tend to forecast better than factor methods and provide an extensive comparison of the strengths and weaknesses of various approaches. Our empirical results show the importance of using forecast metrics which use the entire predictive density, instead of using only point forecasts.

Forecasting and turning point predictions in a Bayesian panel VAR model

We provide methods for forecasting variables and predicting turning points in panel Bayesian VARs. We specify a flexible model which accounts for both interdependencies in the cross section and time variations in the parameters. Posterior distributions for the parameters are obtained for a particular type of diffuse, for Minnesota-type and for hierarchical priors. Formulas for multistep, multiunit point and average forecasts are provided. An application to the problem of forecasting the growth rate of output and of predicting turning points in the G-7 illustrates the approach. A comparison with alternative forecasting methods is also provided.

A Bayesian Approach for Forecasting Heat Load in a District Heating System

The growing population in cities increases the energy demand and affects the environment by increasing carbon emissions. Information and communications technology solutions which enable energy optimization are needed to address this growing energy demand in cities and to reduce carbon emissions. District heating systems optimize the energy production by reusing waste energy with combined heat and power plants. Forecasting the heat load demand in residential buildings assists in optimizing energy production and consumption in a district heating system. However, the presence of a large number of factors such as weather forecast, district heating operational parameters and user behavioural parameters, make heat load forecasting a challenging task. This thesis proposes a probabilistic machine learning model using a Naive Bayes classifier, to forecast the hourly heat load demand for three residential buildings in the city of Skellefteå, Sweden over a period of winter and spring seasons. The district heating data collected from the sensors equipped at the residential buildings in Skellefteå, is utilized to build the Bayesian network to forecast the heat load demand for horizons of 1, 2, 3, 6 and 24 hours. The proposed model is validated by using four cases to study the influence of various parameters on the heat load forecast by carrying out trace driven analysis in Weka and GeNIe. Results show that current heat load consumption and outdoor temperature forecast are the two parameters with most influence on the heat load forecast. The proposed model achieves average accuracies of 81.23 % and 76.74 % for a forecast horizon of 1 hour in the three buildings for winter and spring seasons respectively. The model also achieves an average accuracy of 77.97 % for three buildings across both seasons for the forecast horizon of 1 hour by utilizing only 10 % of the training data. The results indicate that even a simple model like Naive Bayes classifier can forecast the heat load demand by utilizing less training data.

A Bayesian approach for multi-model downscaling: Seasonal forecasting of regional rainfall and river flows in South America

This study addresses three issues: spatial downscaling, calibration, and combination of seasonal predictions produced by different coupled ocean-atmosphere climate models. It examines the feasibility Of using a Bayesian procedure for producing combined, well-calibrated downscaled seasonal rainfall forecasts for two regions in South America and river flow forecasts for the Parana river in the south of Brazil and the Tocantins river in the north of Brazil. These forecasts are important for national electricity generation management and planning. A Bayesian procedure, referred to here as forecast assimilation, is used to combine and calibrate the rainfall predictions produced by three climate models. Forecast assimilation is able to improve the skill of 3-month lead November-December-January multi-model rainfall predictions over the two South American regions. Improvements are noted in forecast seasonal mean values and uncertainty estimates. River flow forecasts are less skilful than rainfall forecasts. This is partially because natural river flow is a derived quantity that is sensitive to hydrological as well as meteorological processes, and to human intervention in the form of reservoir management.

Forecasting with Bayesian multivariate vintage-based VARs

We consider the forecasting of macroeconomic variables that are subject to revisions, using Bayesian vintage-based vector autoregressions. The prior incorporates the belief that, after the first few data releases, subsequent ones are likely to consist of revisions that are largely unpredictable. The Bayesian approach allows the joint modelling of the data revisions of more than one variable, while keeping the concomitant increase in parameter estimation uncertainty manageable. Our model provides markedly more accurate forecasts of post-revision values of inflation than do other models in the literature.

Bayesian space-time data fusion for real-time forecasting and map uncertainty

Environmental computer models are deterministic models devoted to predict several environmental phenomena such as air pollution or meteorological events. Numerical model output is given in terms of averages over grid cells, usually at high spatial and temporal resolution. However, these outputs are often biased with unknown calibration and not equipped with any information about the associated uncertainty. Conversely, data collected at monitoring stations is more accurate since they essentially provide the true levels. Due the leading role played by numerical models, it now important to compare model output with observations. Statistical methods developed to combine numerical model output and station data are usually referred to as data fusion. In this work, we first combine ozone monitoring data with ozone predictions from the Eta-CMAQ air quality model in order to forecast real-time current 8-hour average ozone level defined as the average of the previous four hours, current hour, and predictions for the next three hours. We propose a Bayesian downscaler model based on first differences with a flexible coefficient structure and an efficient computational strategy to fit model parameters. Model validation for the eastern United States shows consequential improvement of our fully inferential approach compared with the current real-time forecasting system. Furthermore, we consider the introduction of temperature data from a weather forecast model into the downscaler, showing improved real-time ozone predictions. Finally, we introduce a hierarchical model to obtain spatially varying uncertainty associated with numerical model output. We show how we can learn about such uncertainty through suitable stochastic data fusion modeling using some external validation data. We illustrate our Bayesian model by providing the uncertainty map associated with a temperature output over the northeastern United States.