Reward prediction errors are quantitative signed terms that express the difference between the value of an obtained outcome and the expected value that was placed on it prior to its receipt. Positive reward prediction errors constitute reward, negative reward prediction errors constitute punishment. Reward prediction errors have been shown to be powerful drivers of reinforcement learning in formal models and there is thus a strong reason to believe they are used in the brain. Isolating such neural signals stands to help elucidate how reinforcement learning is implemented in the brain, and may ultimately shed light on individual differences, psychopathologies of reward such as addiction and depression, and the apparently non-normative behaviour under risk described by behavioural economics. In the present thesis, I used the event related potential technique to isolate and study electrophysiological components whose behaviour resembled reward prediction errors. I demonstrated that a candidate component, “feedback related negativity”, occurring 250 to 350 ms after receipt of reward or punishment, showed such behaviour. A meta-analysis of the existing literature on this component, using a novel technique of “great grand averaging”, supported this view. The component showed marked asymmetries however, being more responsive to reward than punishment and more responsive to appetitive rather than aversive outcomes. I also used novel data-driven techniques to examine activity outside the temporal interval associated with the feedback related negativity. This revealed a later component responding solely to punishments incurred in a Pavlovian learning task. It also revealed numerous salience-encoding components which were sensitive to a prediction error’s size but not its sign.

本研究乃透過國際間各國之比較，以了解匯率變動對貿易收支與國內產生之效果。共 分成兩大部分，第一部分探討匯率變動對貿易收支的影響。根據傳統的國際收支理論 ，認為在某些情況下，一國匯率的貶值可以改善貿易收支。Miles （１９７９）首先 針對匯率與貿易收支的關係進行實證研究。然而，由於生產與消費有時間落後的現象 ，匯率變動對貿易的效果往往無法立即反映，甚至呈現先惡化再逐步改善的情況，一 般稱為Ｊ曲線效果。針對開發中國家是否存在Ｊ曲線效果，Bahmani-Oskooee （１９ ８５）利用Almon 三階多項分配加以驗證。本文係根據他的模型加以延伸，文中運用 Almon 多項分配方法探討１９７３至１９８７年十個工業國家與十個開發中國家，匯 率變動對貿易收支的影響。為了估計落後期數與多項分配階數，使用兩階段的程序， 首先以Akaike提出的final prediction errer為標準，選擇最適落後期數，然後再檢 定多項分配的階數。第二部分探討匯率變動對產出的影響，依照傳統的理論，匯率貶 值會引起出口增加、進口減少，使國內產出擴充。然而Cooper（１９７１）觀察落後 國家資料，卻發現貶值對國內產出有緊縮效果。Edwards （１９８６）首先使用變異 成分分析開發中國家之匯率變動效果，本文對Edwards 的模型加以延伸，利用Almon 多項分配方法估計，檢視匯率變動是否有Cooper效果，同時比較國際間各國之差異。

Master of Science / Department of Agricultural Economics / Glynn Tonsor / The purpose of our analysis was to examine production strategies in the backgrounder and stocker segments of the beef industry within the Flint Hills region of Kansas. The time period analyzed encompassed 1996-2015. September and November placements of steers in the backgrounding sector of the industry were analyzed with an intended March sale date. Placements considered included 425, 500, and 575 pound steers. April and May placements of steers were analyzed for the industry’s stocking sector with an intended July sale date. Placements considered included 450, 600, and 750 pound steers. Within our analysis historical ex-post net incomes were analyzed, prediction errors were calculated (net income, revenue, and cost of gain), and market incentives/signals were analyzed. While for our historical ex-post net income analysis we did not identify one of the four placement strategies as superior in all 20 years of our analysis, we did find scenarios that were typically superior to others. In terms of backgrounding, November placements were typically superior to September placements, in terms of stocking April placements were typically superior to May placements, and when comparing backgrounding and stocking scenarios stocking scenarios were typically superior. In terms of prediction errors, we found that revenue errors are the main drivers of net income error. In general, within the backgrounding scenarios typical producers who are representative of our model assumptions generally overestimate net incomes which is detrimental to them (make lower profits than they anticipate making), while in stocking scenarios producers underestimate net incomes which is generally beneficial to them (make larger profits than they anticipate making). Market signal/incentive and ex-post net income analysis both indicated that steer weight at time of sale was a large factor influencing backgrounder profitability and decision making, and that pasture rents were a large factor influencing stocker profitability and decision making. In all four scenarios it proved economically beneficial to place lighter steer rather than heavy steers. Further research may include, but is not limited to; adding bulls and heifers to our model, analyzing different placement weights within our model, and allowing for animal performance variability within our model.

This diploma thesis describes the Bornhuetter-Ferguson method, which is used to calculate the IBNR reserve. It is divided into deterministic and stochastic parts. The deterministic part deals with the derivation of development pattern and ultimate loss amount, which are needed to calculate the reserve. The stochastic part deals with reserve estimation error and prediction error. The calculation results of the reserve estimate and its error are compared with the results of the chain ladder method. The last chapter deals with the problematic areas of the described method.

<p>In several projects within ABB there is a need of state and parameter estimation for nonlinear dynamic systems. One example is a project investigating optimisation of gas turbine operation. In a gas turbine there are several parameters and states which are not measured, but are crucial for the performance. Such parameters are polytropic efficiencies in compressor and turbine stages, cooling mass flows, friction coefficients and temperatures. Different methods are being tested to solve this problem of system identification or parameter estimation. This thesis describes the implementation of such a method and compares it with previously implemented identification methods. The comparison is carried out in the context of parameter estimation in gas turbine models, a dynamic load model used in power systems as well as models of other dynamic systems. Both simulated and real plant measurements are used in the study.</p>

In several projects within ABB there is a need of state and parameter estimation for nonlinear dynamic systems. One example is a project investigating optimisation of gas turbine operation. In a gas turbine there are several parameters and states which are not measured, but are crucial for the performance. Such parameters are polytropic efficiencies in compressor and turbine stages, cooling mass flows, friction coefficients and temperatures. Different methods are being tested to solve this problem of system identification or parameter estimation. This thesis describes the implementation of such a method and compares it with previously implemented identification methods. The comparison is carried out in the context of parameter estimation in gas turbine models, a dynamic load model used in power systems as well as models of other dynamic systems. Both simulated and real plant measurements are used in the study.

The aim of this thesis is to develop a methodology for obtaining mass properties of a vehicle using specific test rig. Investigated mass properties are the mass, location of center of gravity and the inertia tensor. Accurate measurement of mass properties of vehicles is crucial for vehicle dynamics research. The test rig consists of a frame on which the vehicle is fixed and which is suspended from the ceiling of the laboratory using steel cables. Mass and location of center of gravity are measured using the data from the test rig in equilibrium position and basic static equations. Inertia tensor is measured using the data from dynamical response of the system. For this purpose an identification routine which employs prediction error method is developed using the built&ndash / in functions from the System Identification Toolbox of MATLAB&reg / . The experiment was also simulated using Simmechanics Toolbox of MATLAB&reg / . Identification code is verified using the results of the experiment simulations for various cases.

Όπως δηλώνει και ο τίτλος, η παρούσα διπλωματική εργασία διαπραγματεύεται διάφορες τεχνικές για την ανίχνευση του φάσματος σε cognitive radio SIMO συστήματα. Η συμβατική προσέγγιση της διαχείρισης του φάσματος δεν είναι ευέλικτη καθώς με το περισσότερο χρήσιμο τμήμα του ραδιοφάσματος να είναι δεσμευμένο, είναι εξαιρετικά δύσκολο να βρεθούν ελεύθερες συχνότητες για την ανάπτυξη νέων υπηρεσιών ή για τον εμπλουτισμό των ήδη υπαρχόντων, ενώ ταυτόχρονα, διάφορες μετρήσεις έχουν καταδείξει ότι το αδειοδοτημένο φάσμα σπάνια χρησιμοποιείται πλήρως, τόσο ως προς το πεδίο του χρόνου όσο και ως προς το πεδίο του χώρου. Έτσι, η τεχνολογία του Cognitive Radio (Γνωστικά Συστήματα Ραδιοεπικοινωνιών) έρχεται να προσφέρει λύση, κυρίως, στα παραπάνω ζητήματα παρέχοντας δυναμική εκμετάλλευση του φάσματος. Η τεχνολογία του Cognitive Radio έχει προταθεί για μικρότερης προτεραιότητας δευτερεύοντα συστήματα αποσκοπώντας στη βελτίωση της αποδοτικότητας του διαθέσιμου φάσματος μέσω της ανίχνευσής του και επιτρέποντας στα δευτερεύοντα αυτά συστήματα να εκπέμπουν στις μπάντες που εντοπίζονται να μη χρησιμοποιούνται. Όπως γίνεται εύκολα αντιληπτό από τα παραπάνω, η ανίχνευση φάσματος (spectrum sensing) αποτελεί ένα ιδιαιτέρως κρίσιμο θέμα για τα cognitive συστήματα. Για να επιτευχθεί η προσαρμοστική μετάδοση σε αχρησιμοποίητα τμήματα φάσματος, χωρίς να προκαλούνται παρεμβολές στους βασικούς χρήστες αυτών των τμημάτων (Primary Users-PUs), το spectrum sensing αποτελεί το πρώτο και ένα από τα κυριότερα βήματα, καθώς απαιτείται υψηλή αξιοπιστία στην ανίχνευση του σήματος των PUs. Οι δευτερεύοντες χρήστες (Secondary Users-SUs), δηλαδή, θα πρέπει να γνωρίζουν αν το φάσμα χρησιμοποιείται ώστε να αξιοποιήσουν το διαθέσιμο φάσμα με τον πιο αποτελεσματικό τρόπο. Ουσιαστικά, το spectrum sensing εφαρμόζεται για να δώσει στον cognitive χρήστη μια όσο το δυνατόν πιστότερη εικόνα του περιβάλλοντος στο οποίο βρίσκεται. Σκοπό της παρούσας διπλωματικής εργασίας αποτελεί η μελέτη και η ανάπτυξη αλγορίθμων που θα επιτρέπουν στον SU ενός SIMO συστήματος να ανιχνεύει την ύπαρξη φασματικών κενών. Η υλοποίηση που χρησιμοποιήσαμε βασίζεται στη χρήση ενός predictor. Πιο συγκεκριμένα, το σήμα που λαμβάνει ο δέκτης περνά από ένα backward linear predictor από τον οποίο υπολογίζουμε τη διαφορά του προβλεπόμενου σήματος σε σχέση με το πραγματικό, δηλαδή το σφάλμα πρόβλεψης. Αξιοποιώντας κατάλληλα το σφάλμα πρόβλεψης, και πιο συγκεκριμένα τον πίνακα αυτοσυσχέτισης του σφάλματος, μας δίνεται η δυνατότητα να ανιχνεύσουμε αξιόπιστα την ύπαρξη ή την απουσία σήματος, ακόμα και σε θορυβώδη περιβάλλοντα, δηλαδή για χαμηλές τιμές του λόγου σήματος προς θόρυβο. Για τον έλεγχο της απόδοσης των αλγορίθμων που αναπτύξαμε, το παραπάνω σύστημα εξομοιώθηκε σε MATLAB για διάφορες συνθήκες και κανάλια / In the present thesis, we will study spectrum sensing techniques of Cognitive Radio SIMO systems. The conventional approach to spectrum management is not flexible, as most of the useful part of the spectrum is bounded. Hence it is extremely difficult to find free frequencies in order to deploy new services or to enhance the already existing ones. At the same time, various measurements show that the licensed spectrum is heavily underutilized in terms of both the time domain as well as the space domain. Thus Cognitive Radio technology comes to offer solutions, mainly with regard to the issues mentioned above, providing a dynamic utilization of the spectrum. Cognitive Radio has been proposed for lower priority secondary systems intending to improve spectral efficiency through spectrum sensing thus allowing these systems to transmit at frequency bands that are detected to be unused. As we can easily understand from the above, spectrum sensing is a critical issue for cognitive systems. In order to achieve adaptive transmission in unused portions of the spectrum without interferences to the licensed users of these portions (Primary Users-PUs), spectrum sensing is the first and one of the most important steps as high reliability is demanded on PUs' signal detection. That is, Secondary Users (SUs) should know if the spectrum is being used in order to exploit the available spectrum in the most efficient way. Essentially, spectrum sensing is used in order to provide the cognitive user with a representation of its operating environment which is as faithful as possible. The scope of this thesis is the study and the creation of algorithms that will give the SU of a SIMO system the opportunity to detect the existence of spectrum holes. The implementation we used is based on a predictor. More specifically, the received signal passes through a backward linear predictor from which we compute the difference between the actual signal and the predicted signal, which is the prediction error. By properly exploiting the prediction error, more precisely the autocorrelation matrix of the prediction error, we can trustworthily detect the existence or the absence of a signal, even in noisy environments, that is, for low values of the signal-to-noise ratio. In order to test the performance of our algorithms, the system above was simulated by MATLAB for different conditions and channels.

In modern manufacturing, industrial robots are essential components that allow saving cost, increase quality and productivity for instance. To achieve such goals, high accuracy and speed are simultaneously required. The design of control laws compliant with such requirements demands high-fidelity mathematical models of those robots. For this purpose, dynamic models are built from experimental data. The main objective of this thesis is thus to provide robotic engineers with automatic tools for identifying dynamic models of industrial robot arms. To achieve this aim, a comparative analysis of the existing methods dealing with robot identification is made. That allows discerning the advantages and the limitations of each method. From those observations, contributions are presented on three axes. First, the study focuses on the estimation of the joint velocities and accelerations from the measured position, which is required for the model construction. The usual method is based on a home-made prefiltering process that needs a reliable knowledge of the system’s bandwidths, whereas the system is still unknown. To overcome this dilemma, we propose a method able to estimate the joint derivatives automatically, without any setting from the user. The second axis is dedicated to the identification of the controller. For the vast majority of the method its knowledge is indeed required. Unfortunately, for copyright reasons, that is not always available to the user. To deal with this issue, two methods are suggested. Their basic philosophy is to identify the control law in a first step before identifying the dynamic model of the robot in a second one. The first method consists in identifying the control law in a parametric way, whereas the second one relies on a non-parametric identification. Finally, the third axis deals with the home-made setting of the decimate filter. The identification of the noise filter is introduced similarly to methods developed in the system identification community. This allows estimating automatically the dynamic parameters with low covariance and it brings some information about the noise circulation through the closed-loop system. All the proposed methodologies are validated on an industrial robot with 6 degrees of freedom. Perspectives are outlined for future developments on robotic systems identification and other complex problems.

Closed-loop system identification and validation are important components in dynamic system modelling. In this dissertation, a comprehensive literature survey is compiled on system identification with a specific focus on closed-loop system identification and issues of identification experiment design and model validation. This is followed by simulated experiments on known linear and non-linear systems and experiments on a pilot scale distillation column. The aim of these experiments is to study several sensitivities between identification experiment variables and the consequent accuracy of identified models and discrimination capacity of validation sets given open and closed-loop conditions. The identified model structure was limited to an ARX structure and the parameter estimation method to the prediction error method. The identification and validation experiments provided the following findings regarding the effects of different feedback conditions: <ul> <li>Models obtained from open-loop experiments produced the most accurate responses when approximating the linear system. When approximating the non-linear system, models obtained from closed-loop experiments were found to produce the most accurate responses.</li> <li>Validation sets obtained from open-loop experiments were found to be most effective in discriminating between models approximating the linear system while the same may be said of validation sets obtained from closed-loop experiments for the nonlinear system.</li> </ul> These finding were mostly attributed to the condition that open-loop experiments produce more informative data than closed-loop experiments given no constraints are imposed on system outputs. In the case that system output constraints are imposed, closed-loop experiments produce the more informative data of the two. In identifying the non-linear system and the distillation column it was established that defining a clear output range, and consequently a region of dynamics to be identified, is very important when identifying linear approximations of non-linear systems. Thus, since closed-loop experiments produce more informative data given output constraints, the closed-loop experiments were more effective on the non-liner systems. Assessment into other identification experiment variables revealed the following: <ul> <li>Pseudo-random binary signals were the most persistently exciting signals as they were most consistent in producing models with accurate responses.</li> <li>Dither signals with frequency characteristics based on the system’s dominant dynamics produced models with more accurate responses.</li> <li>Setpoint changes were found to be very important in maximising the generation of informative data for closed-loop experiments</li></ul> Studying the literature surveyed and the results obtained from the identification and validation experiments it is recommended that, when identifying linear models approximating a linear system and validating such models, open-loop experiments should be used to produce data for identification and cross-validation. When identifying linear approximations of a non-linear system, defining a clear output range and region of dynamics is essential and should be coupled with closed-loop experiments to generate data for identification and cross-validation. / Dissertation (MEng)--University of Pretoria, 2009. / Chemical Engineering / unrestricted