Graduation date: 2006 / Regression calibration inference seeks to estimate regression models with measurement error in explanatory variables by replacing the mismeasured variable by its conditional expectation, given a surrogate variable, in an estimation procedure that would have been used if the true variable were available. This study examines the effect of the uncertainty in the estimation of the required conditional expectation on inference about regression parameters, when the true explanatory variable and its surrogate are observed in a calibration dataset and related through a normal linear model. The exact sampling distribution of the regression calibration estimator is derived for normal linear regression when independent calibration data are available. The sampling distribution is skewed and its moments are not defined, but its median is the parameter of interest. It is shown that, when all random variables are normally distributed, the regression calibration estimator is equivalent to maximum likelihood provided a natural estimate of variance is non-negative. A check for this equivalence is useful in practice for judging the suitability of regression calibration. Results about relative efficiency are provided for both external and internal calibration data. In some cases maximum likelihood is substantially more efficient than regression calibration. In general, though, a more important concern when the necessary conditional expectation is uncertain, is that inferences based on approximate normality and estimated standard errors may be misleading. Bootstrap and likelihood-ratio inferences are preferable.

We construct polar codes for binary relay channels with orthogonal receiver components. We show that polar codes achieve the cut-set bound when the channels are symmetric and the relay-destination link supports compress-and-forward relaying based on Slepian-Wolf coding. More generally, we show that a particular version of the compress-and-forward rate is achievable using polar codes for Wyner-Ziv coding. In both cases the block error probability can be bounded as O(2-Nβ) for 0 &lt; β &lt; 1/2 and sufficiently large block length N. / <p>© 2010 IEEE. Personal use of this material is permitted. Permission from IEEE must be obtained for all other uses, in any current or future media, including reprinting/republishing this material for advertising or promotional purposes, creating new collective works, for resale or redistribution to servers or lists, or reuse of any copyrighted component of this work in other works. QC 20111207</p>

The following thesis is a project in mathematical biology building upon the so-called "error minimization hypothesis" of the genetic code. After introducing the biological context of this hypothesis, I proceed to develop some relevant information-theoretic ideas, with the overall goal of studying the structure of the genetic code. I then apply the newfound understanding to an important question in the debate about the origin of life, namely, the question of the temperatures in which the genetic code, and life in general, underwent their early evolution. <br /><br /> The main advance in this thesis is a set of methods for calculating the primordial evolutionary pressures that shaped the genetic code. These pressures are due to genetic errors, and hence the statistical properties of the errors and of the genome are imprinted in the statistical properties of the code. Thus, by studying the code it is possible to reconstruct, to some extent, the primordial error rates and the composition of the primordial genome. In this way, I find evidence that the fixation of the genetic code occurred in organisms which were not thermophiles.

Statistical models are often used for the prediction of future random variables. There are two types of prediction, point prediction and probabilistic prediction. The prediction accuracy is quantified by performance measures, which are typically based on loss functions. We study the estimators of these performance measures, the prediction error and performance scores, for point and probabilistic predictors, respectively. The focus of this thesis is to assess the prediction performance of survival models that analyze censored survival times. To accommodate censoring, we extend the inverse probability censoring weighting (IPCW) method, thus arbitrary loss functions can be handled. We also develop confidence interval procedures for these performance measures. We compare model-based, apparent loss based and cross-validation estimators of prediction error under model misspecification and variable selection, for absolute relative error loss (in chapter 3) and misclassification error loss (in chapter 4). Simulation results indicate that cross-validation procedures typically produce reliable point estimates and confidence intervals, whereas model-based estimates are often sensitive to model misspecification. The methods are illustrated for two medical contexts in chapter 5. The apparent loss based and cross-validation estimators of performance scores for probabilistic predictor are discussed and illustrated with an example in chapter 6. We also make connections for performance.

The following thesis is a project in mathematical biology building upon the so-called "error minimization hypothesis" of the genetic code. After introducing the biological context of this hypothesis, I proceed to develop some relevant information-theoretic ideas, with the overall goal of studying the structure of the genetic code. I then apply the newfound understanding to an important question in the debate about the origin of life, namely, the question of the temperatures in which the genetic code, and life in general, underwent their early evolution. <br /><br /> The main advance in this thesis is a set of methods for calculating the primordial evolutionary pressures that shaped the genetic code. These pressures are due to genetic errors, and hence the statistical properties of the errors and of the genome are imprinted in the statistical properties of the code. Thus, by studying the code it is possible to reconstruct, to some extent, the primordial error rates and the composition of the primordial genome. In this way, I find evidence that the fixation of the genetic code occurred in organisms which were not thermophiles.

Statistical models are often used for the prediction of future random variables. There are two types of prediction, point prediction and probabilistic prediction. The prediction accuracy is quantified by performance measures, which are typically based on loss functions. We study the estimators of these performance measures, the prediction error and performance scores, for point and probabilistic predictors, respectively. The focus of this thesis is to assess the prediction performance of survival models that analyze censored survival times. To accommodate censoring, we extend the inverse probability censoring weighting (IPCW) method, thus arbitrary loss functions can be handled. We also develop confidence interval procedures for these performance measures. We compare model-based, apparent loss based and cross-validation estimators of prediction error under model misspecification and variable selection, for absolute relative error loss (in chapter 3) and misclassification error loss (in chapter 4). Simulation results indicate that cross-validation procedures typically produce reliable point estimates and confidence intervals, whereas model-based estimates are often sensitive to model misspecification. The methods are illustrated for two medical contexts in chapter 5. The apparent loss based and cross-validation estimators of performance scores for probabilistic predictor are discussed and illustrated with an example in chapter 6. We also make connections for performance.

Syfte: Att undersöka sjuksköterskors uppfattningar om avvikelser i vården och om bakomliggande orsaker till dessa samt inställningar till avvikelserapportering och -hantering. Studien syftade även till att undersöka uppfattningar om hur nyanställda sjuksköterskor påverkas av rådande inställningar till detta på en vårdavdelning. Metod: Kvalitativ metod med explorativ design. Bekvämlighetsurval tillämpades och data samlades in med hjälp av semistrukturerade intervjuer med sex sjuksköterskor från en medicinsk vårdavdelning på ett svenskt universitetssjukhus. Dataanalysen genomfördes med fenomenografisk metod. Resultat: Sjuksköterskornas generella uppfattningar om förekommande avvikelser inom vården kan sammanfattas som säkerhetsbrister för patienter och personal. Orsakerna till avvikelser kan indelas i flera kategorier varav ”hög arbetsbelastning” är den mest omfattande. Närapå samtliga orsaker kan härledas till brister i organisation och/eller ledning. Att avvikelser inträffar anses allvarligt men mänskligt och ofrånkomligt. Sjuksköterskorna har olika attityder till avvikelserapportering t.ex. att man ser det som ett led i förbättringsarbetet eller utser syndabockar. Generellt uppfattas att avvikelser underrapporteras. Ett antal faktorer såsom graden av säkerhetsrisk, tidsbrist samt tvivel på uppföljning uppfattas som avgörande för om man väljer att rapportera en avvikelse eller inte. Sjuksköterskorna efterfrågar feedback och information om rapporterade avvikelser. Nyanställda sjuksköterskor uppfattas påverkas starkt av andra sjuksköterskor gällande attityder till avvikelser såväl som avvikelserapportering. / Aim: To examine nurses' opinion about errors in their profession and the underlying causes of these. A further aim was to study attitudes towards error reporting and to examine whether more experienced colleagues’ opinion concerning error making and error reporting had any effect on newly employed inexperienced nurses. Methods: The study was qualitative with an explorative design. Convenience sampling was applied and data was collected by semi-structural interviews. Six nurses from a medical care unit participated in the study and data was analyzed with a phenomenographic method. Results: The major category of nurses' views about error making included lack of patient safety and working environment issues is another category. It could all be summed up in one major category called ”lack of safety” including both patients and personnel. Several categories were identified as causes of error making, among those high workload is the most frequently mentioned cause. Nearly all of the causes can be traced to lacking in organization and/or management. The fact that errors occur are regarded as severe but human and inevitable. Several attitudes towards reporting errors were emerged from the data, for example regarding it as a tool of improvement work or finding scapegoats. The most common view was that errors are underreported and a number of factors regarding nurses' willingness to report the errors were found, for example severeness of the error, lack of time and doubt in follow up. Nurses wished to get feedback and information about reported errors. Inexperienced nurses were highly influenced by their senior colleagues concerning both error making and error reporting.

The present study investigates three teacher trainees and three teachers’ views on error correction during oral communication, and the similarities and differences between them. These six people were interviewed separately and they were asked six questions; the first five questions were asked to all six people but the last question differed between the teacher trainees and the teachers. My result shows that the teacher trainees are insecure when it comes to error correction and that the teachers´ sees it as a part of their job, and that is the biggest difference between them. The teacher trainees and the teachers focus on the same types of errors and those are the errors that can cause problems in communication, and that can be pronunciation errors, grammatical errors or vocabulary errors.

This paper discusses the factors that affect the share price index of Taiwan solar power industry, crude oil prices and the size of the solar manufacturers in Taiwan and Taiwan's market index into the consideration. In addition, considering whether the policies implemented by our government would change the solar industry in Taiwan¡¦s stocks structural .Using the correlation coefficient, the unit root test, Chow test, cointegration test , vector error correction model, impulse response and forecast error variance decomposition to explore their relationship respectively. The study period starts from January 3,2002 until December 30,2011, a total of 2450 daily data for empirical analysis. By Chow test , we find that there is no structural change of solar stock index after the implementation of the domestic policies. Three international crude oil prices and the total share capital of solar manufacturers in Taiwan and the Taiwan solar power industry stocks index has co-integration relationship, means the three international crude oil prices and solar companies total share capital of solar stock index has a long-run equilibrium relationship. By the error correction model of West Texas crude oil price of Brent crude oil prices, the total share capital of the solar companies in Taiwan and Taiwan solar stock index mutual interaction, and the relationship between changes in Taiwan's solar stock price index and Brent crude oil price, West Texas crude oil prices and the total manufacturers of solar energy manufacturers in Taiwan¡¦s share capital are positive, besides, I also found a positive relationship in the impulse response.

Recently, interference alignment (IA) has emerged as a promising technique to effectively mitigate interference in wireless communication systems. It has also evolved as a powerful technique to achieve the optimal degrees of freedom of interference channel. IA can be constructed in many domains such as space, time, frequency and codes. Currently, most researches on developing IA assume that channel state information (CSI) is well-known at the transceiver. However, in practice, perfect CSI at the transceiver can¡¦t be obtained due to many factors such as channel estimation error, quantization error, and feedback error. Under our investigation, the performance of IA is very sensitive to imperfect CSI. Therefore, this thesis proposes a spatial domain IA scheme for the three-user multiple-input multiple-output (MIMO) downlink interference channels, and analyzes the effect of channel estimation errors by modeling the estimation error as independent complex Gaussian random variables. The approximated bit error rate (BER) for the system with MIMO Zero-Forcing equalizer using IA is derived.