This is an empirical study based on English as a foreign language (EFL) teachers' experiences. The study is on Chinese EFL teachers’ usage of different oral corrective feedback (OCF) types as well as their own beliefs as to what strategies they make use of most frequently and what they do to improve their own OCF. Semi-structured interviews and non-participant observations were employed as instruments for the data collecting process. Four Chinese EFL teachers were interviewed, one male and three female, with varied amounts of professional experience and of different ages. Four observations were made in the classrooms of the same teachers as a supplement to the interviews. The interviews were analyzed by means of content analysis. The results indicate that Recast is the most commonly used feedback method and that the teachers had different thoughts on which feedback method was the most effective or the one they most frequently used. Furthermore, the results indicate that there are differences between teachers’ beliefs about their feedback strategies and the observation results. / Detta är en empirisk studie som bygger på intervjuer med och observationer av kinesiska lärare i engelska som främmande språk (EFS). Målet med studien var att få en uppfattning om kinesiska EFS-lärares användning av olika muntliga korrigerande feedbackmetoder (MKF) samt deras egna föreställningar om vilka strategier som de använder oftast och vad de strävar efter att förbättra i sina egna sätt att ge MKF. Semistrukturerade intervjuer och ickedeltagande observation användes som instrument under datainsamlingsprocessen. Fyra kinesiska EFS-lärare intervjuades, en manlig och tre kvinnliga lärare med varierande längd av yrkeserfarenhet och i olika åldrar. Fyra observationer gjordes även på samma lärares lektioner som ett komplement till intervjuerna. Intervjuerna analyserades utifrån metoder för innehållsanalys. Resultaten indikerar att omformulering är den mest använda feedbackmetoden och att lärarna hade delade åsikter om vilken feedbackmetod som är mest effektiv eller den som de själva använde mest. Vidare tyder resultaten på att det finns skillnader mellan lärarnas föreställningar om sina feedbackstrategier och observationsresultaten.

The aim of this diploma thesis is to analyze the development of the total volume of debt in the Czech Republic and the analysis separately for the household sector and non-financial corporations. From economic theoretical assumptions it can be concluded that there is a correlation between the amount of loans and GDP development or between credit and economic cycle. The thesis is divided into three parts. The first part made up of chapters 1 to 4, describes the theory used further in the text. The second part, Chapter 5, describes the specific time series used in the thesis, i.e. The time series of the volume of debt for the Czech Republic, GDP and interest rates. Interest rates and the volume of debt are further broken down by maturity and also by two selected sectors. The last part, Chapter 6, focusing on co-integration analysis, ADL and error correction models, attempts to capture short-term and long-term relationships between the time series.

Please read the abstract in the section 00front of this document. / Dissertation (MEng (Electronic Engineering))--University of Pretoria, 2006. / Electrical, Electronic and Computer Engineering / unrestricted

In practice, most approaches for text error detection and correction are based on a conventional domain-dependent background dictionary that represents a fixed and static collection of correct words of a given language and, as a result, satisfactory correction can only be achieved if the dictionary covers most tokens of the underlying correct text. Again, most approaches for text correction are for only one or at best a very few types of errors. The purpose of this thesis is to propose an unsupervised approach to detecting and correcting text errors, that can compete with supervised approaches and answer the following questions: Can an unsupervised approach efficiently detect and correct a text containing multiple errors of both syntactic and semantic nature? What is the magnitude of error coverage, in terms of the number of errors that can be corrected? We conclude that (1) it is possible that an unsupervised approach can efficiently detect and correct a text containing multiple errors of both syntactic and semantic nature. Error types include: real-word spelling errors, typographical errors, lexical choice errors, unwanted words, missing words, prepositional errors, article errors, punctuation errors, and many of the grammatical errors (e.g., errors in agreement and verb formation). (2) The magnitude of error coverage, in terms of the number of errors that can be corrected, is almost double of the number of correct words of the text. Although this is not the upper limit, this is what is practically feasible. We use engineering approaches to answer the first question and theoretical approaches to answer and support the second question. We show that finding inherent properties of a correct text using a corpus in the form of an n-gram data set is more appropriate and practical than using other approaches to detecting and correcting errors. Instead of using rule-based approaches and dictionaries, we argue that a corpus can effectively be used to infer the properties of these types of errors, and to detect and correct these errors. We test the robustness of the proposed approach separately for some individual error types, and then for all types of errors. The approach is language-independent, it can be applied to other languages, as long as n-grams are available. The results of this thesis thus suggest that unsupervised approaches, which are often dismissed in favor of supervised ones in the context of many Natural Language Processing (NLP) related tasks, may present an interesting array of NLP-related problem solving strengths.

The degree work is focused on analysis of inflation with help of suitable econometric models. Inflation with it's forms and possibilities of measuring is described at the beginning of the paper. There is mentioned an importance of monitoring and analysing inflation in view of Czech national bank. Consequently there are described characteristics of time series, which are important from viewpoint of construction of econometric models. Next part of this paper is focused on characterization of econometrics models. At first there is vector autoregression model, in this connection there is discussed the essence of Granger causality and impulse reaction. There are also noticed both error correction model and vector error correction model. The empirical part of degree work involves the use of these models on selected macroeconomic time series of the Czech republic. The objective is to analyze the relationship between inflation and other individual macroeconomic quantities. There is established the optimal vector autoregressive model and the results of Granger causality and impulse reaction are interpretated. Both error correction model and vector error correction model examining cointegration are also applied.

This work deals with the demand for money theories and applications of error correction Models and VAR models for the demand for money in the Czech Republic. In the theoretical part of the work are discussed theory of demand for money, with an emphasis on Patinkins theory of demand for money and the wording of VAR models and error correction models. In the application part are applied VAR (1), VAR (2) models and error correction models on the demand for money in the Czech Republic. Data used in the work come from the information system CNB and the CZSO.

This paper examines the (in contemporary literature inconclusive) usefulness ofcointegration between stock prices as basis for a trading strategy. The primary contributionto previously used frameworks of the paper is the implementation and use of error correctionmodels for selection of stocks to trade on. Evaluation is done through simulated resultsrunning the algorithm on the sectors of the Standard and Poor’s 500 index in the years 2005through 2014. Results indicate that trading strategies of this nature may be very successfuleven in recent years given that the universe of tradeable stocks within a sector is sufficientlylarge. The application of error correction models improve average returns, though in a waynot originally anticipated.

I experimented with Hopfield networks in the context of a voice-based, query-answering system. Hopfield networks are used to store and retrieve patterns. I used this technique to store queries represented as natural language sentences and I evaluated the accuracy of the technique for error correction in a spoken question-answering dialog between a computer and a user. I show that the use of an auto-associative Hopfield network helps make the speech recognition system more fault tolerant. I also looked at the available encoding schemes to convert a natural language sentence into a pattern of zeroes and ones that can be stored in the Hopfield network reliably, and I suggest scalable data representations which allow storing a large number of queries.

The education community agrees that correcting student errors is important for learning. They do not agree on the components that define successful error correcting. Some theories suggest that detailed feedback facilitates adult learning and some suggest that less detail is needed for these learners. Gilbert (1962) applied the scientifically derived methods of Behavior Analysis when designing instruction. This study attempted to develop an efficient error correction procedure for university teachers. Throughout the semester, error correction design efforts between the teachers and the experimenter became more collaborative. While error correction procedures never showed systematic effects on student grades, later versions were viewed more favorably by both teachers and students and were more likely to be implemented accurately. Decreased teacher practice opportunities, due to low student participation, may have decreased the procedure's effectiveness.

O mercado do milho no Brasil tem demonstrado algumas mudanças nos últimos anos aumentando sua importância no agronegócio. Por esse motivo, tem sido maior a necessidade de estudo de suas características. O objetivo da presente dissertação consiste em verificar como se dá a transmissão de preços entre os mercados regionais dessa commodity no Brasil a nível de produtor, com foco nos estados da região Sul e Centro-Oeste, devido à sua importância para a produção nacional. Além disso, também foi buscado analisar como os preços dos estados analisados reagem ao preço cotado na bolsa de valores para saber qual sua relação com o mercado externo. O método de pesquisa utilizado foi: teste de raiz unitária; teste de cointegração; vetor de correção de erro; teste de causalidade de Granger e teste de impulso-resposta. Os resultados do teste de cointegração indicam que há transmissão de preços entre todos os estados analisados, bem como os estados respondem a oscilações de preços do mercado externo a longo prazo. O fato de existir cointegração entre os estados é condição suficiente para se afirmar que existe relação linear de equilíbrio para a qual o sistema converge, validando os pressupostos da Lei do Preço Único e a integração. Todos os estados apresentaram resposta significativa a mudanças de preços no estado de Santa Catarina pelo vetor de correção de erro (VEC), mostrando que esse estado tem forte influência na formação de preços dos estados das duas regiões analisadas. No curto prazo foi observado que os estados de Mato Grosso e Rio Grande do Sul não recebem influência direta das oscilações de preços dos outros mercados, enquanto os estados de Paraná, Santa Catarina e Goiás parecem ser interdependentes a curto prazo, pois apresentam relativa correlação. Além disso, a função impulso resposta demonstra também que um impulso nos preços do estado de Santa Catarina gera resposta significativa nos preços dos outros estados de forma geral, e um impulso no preço do estado de Goiás também gera uma reação forte no preço do estado do Mato Grosso. / The corn market in Brazil has shown some changes in recent years increasing its importance in agribusiness. For this reason has increased the need for more studies related to this market’s characteristics . The objective of this dissertation is to check how is the price transmission between regional markets in Brazil at producer level for this commodity, focusing on states of the South and Midwest of the country, because of its importance to the national production. Furthermore, it was also sought to analyze how the prices of the analyzed states react to the price quoted on the stock market, looking for understanding how is its relationship with the external market. The research method used was: the unit root test , cointegration test , vector error correction; Granger causality test and impulse response test. The result of the cointegration test indicates that there is price transmission between all the states analyzed as well as states respond to price fluctuations on the stock market in the long run . The existence of cointegration between the states is sufficient to say that there is a linear equilibrium relationship to which converges the sistem, validating the assumptions of the Law of One Price and the integration condition. All states showed significant responses to price changes in the state of Santa Catarina by the vector error correction ( VEC ) , showing that this state has a strong influence on the pricing of the states on the two regions. In the short term it was observed that the states of Mato Grosso and Rio Grande do Sul receive no direct influence from the prices of other markets, while the states of Paraná, Santa Catarina and Goiás seem to be interdependent in the short term because they present a correlation. As well, the impulse response function also shows that a surge in prices in the state of Santa Catarina generates a significant response in prices of other states in general, and a boost in the price of Goias also generates a strong reaction in the price of Mato Grosso.