This study examines the efficacy of presenting footnotes information in alternative display formats on investors’ judgments and decisions. Non-professional investors play a significant role in the capital markets yet they do not always attend to information contained in footnote disclosures. As a result, nonprofessional investors systematically misprice firms and misallocate resources. Recognizing that increased mandatory and voluntary disclosures create additional challenges for non-professional investors, both the FASB and SEC have actively sought ways to increase the effectiveness of disclosures. I hypothesize that high display proximity, low signal-to-noise presentation formats can increase investors’ attention to and processing of footnote disclosures and hence performance on an investing task. Further I hypothesize that low display proximity, low signal-to-noise presentation formats can improve investor performance on a recognition task. Lastly, I hypothesize that non-professional investors viewing high display proximity, low signal-to-noise footnote disclosures will rate usability higher than nonprofessional investors viewing footnote disclosures in the other three display formats. Amazon Mechanical Turk workers are used as participants in a 2 x 2 between-participants experiment using two task types: an integrative (investing) task and a non-integrative (recognition) task. I manipulate display proximity (inline or side-by-side) and signal-to-noise ratio (footnotes presented simultaneously or individually). Contrary to my hypotheses, I find that low signal-to-noise ratio increases non-professional investors’ performance on both the integrative (investing) and non-integrative (recognition) tasks. Further, although task performance increased under the low signal-to-noise presentation format, participants did not evaluate either signal-to-noise presentation format easier to use or more useful. Instead, participants found the high display proximity (side-by-side) presentation format easier to use, although it did not yield performance increases.

Orientadores: Hugo Enrique Hernandez Figueroa, Júlio César Rodrigues Fernandes de Oliveira / Dissertação (mestrado) - Universidade Estadual de Campinas, Faculdade de Engenharia Elétrica e de Computação / Made available in DSpace on 2018-08-26T11:25:22Z (GMT). No. of bitstreams: 1 Januario_JoaoCarlosSorianoSampaio_M.pdf: 2247178 bytes, checksum: 866411501272be6234f26634c08a5397 (MD5) Previous issue date: 2014 / Resumo: O alto crescimento da demanda por tráfego nas redes ópticas de comunicações impulsionaram sua rápida evolução a fim de que esta passasse a suportar altas taxas de transmissão e otimizasse os recursos presentes em sua camada física. Para acompanhar este desenvolvimento as arquiteturas de rede também se modernizaram, partindo das redes ponto a ponto para as atuais redes em malha, as quais requerem que uma monitoração da performance do sinal óptico seja realizada ao longo de todo o seu enlace para que os parâmetros adquiridos sirvam de entrada para um controle capaz de proporcionar a flexibilidade que a rede necessita. Por esta razão este trabalho apresenta uma proposta de monitorador da relação sinal ruído óptica (Optical Signal to Noise Ratio, OSNR), principal parâmetro associado à qualidade do sinal em um enlace óptico, baseado nas propriedades de um interferômetro Mach Zehnder. Análises em simulação e experimentais foram conduzidas para avaliar a robustez do método na presença de efeitos dispersivos e verificou-se sua validade tanto para sinais não coerentes modulados em apenas uma polarização quanto para sinais codificados a partir da multiplexação de polarizações. Experimentalmente esta afirmação se confirma através da obtenção de um erro de monitoração menor que 1,5 dB para uma faixa de OSNR de 5 dB à 25 dB para o sinal OOK à 10 Gb/s e uma erro de menor que 3,0 dB para os sinais DP-QPSK à 112 Gb/s e 16QAM à 224 Gb/s para a mesma faixa de variação de OSNR / Abstract: The increasing demand for data in an optical network boosted its fast development in order to allow higher transmission rates and ensure a new generation of components on network physical layer. To follow this development the network architectures evolved too, starting from point to point network and going to actual mesh scenario, which requires a performance monitoring to guarantee a high level of reconfigurability and flexibility. Because of that, this work shows a new approach based on a Mach Zehnder interferometer to monitor optical signal to noise ratio (OSNR), the main parameter related to signal quality in an optical link. Analysis in a simulation environment and experimentally were made to evaluate the robustness of this new method in presence of dispersive effects. With that, it was validated its accuracy for non coherent signals modulated with only one polarization and for signals which exploit polarization multiplexing. Experimental analysis show a monitoring error below 1.5dB for a signal OOK at 10Gb/s with an OSNR range from 5dB to 25dB and for a signal DP-QPSK at 112Bb/s and 16QAM at 224Gb/s a monitoring error below 3.0dB was reached with the same OSNR range / Mestrado / Telecomunicações e Telemática / Mestre em Engenharia Elétrica

No description available.

Background: Since the 1940s, measures of pure-tone sensitivity and speech recognition in quiet have been vital components of the audiologic evaluation. Although early investigators urged that speech recognition in noise also should be a component of the audiologic evaluation, only recently has this suggestion started to become a reality. This report focuses on the Words-in-Noise (WIN) Test, which evaluates word recognition in multitalker babble at seven signal-to-noise ratios and uses the 50% correct point (in dB SNR) calculated with the Spearman-Kärber equation as the primary metric. The WIN was developed and validated in a series of 12 laboratory studies. The current study examined the effectiveness of the WIN materials for measuring the word-recognition performance of patients in a typical clinical setting. Purpose: To examine the relations among three audiometric measures including pure-tone thresholds, word-recognition performances in quiet, and word-recognition performances in multitalker babble for veterans seeking remediation for their hearing loss. Research Design: Retrospective, descriptive. Study Sample: The participants were 3430 veterans who for the most part were evaluated consecutively in the Audiology Clinic at the VA Medical Center, Mountain Home, Tennessee. The mean age was 62.3 yr (SD = 12.8 yr). Data Collection and Analysis: The data were collected in the course of a 60 min routine audiologic evaluation. A history, otoscopy, and aural-acoustic immittance measures also were included in the clinic protocol but were not evaluated in this report. Results: Overall, the 1000-8000 Hz thresholds were significantly lower (better) in the right ear (RE) than in the left ear (LE). There was a direct relation between age and the pure-tone thresholds, with greater change across age in the high frequencies than in the low frequencies. Notched audiograms at 4000 Hz were observed in at least one ear in 41% of the participants with more unilateral than bilateral notches. Normal pure-tone thresholds (≤20 dB HL) were obtained from 6% of the participants. Maximum performance on the Northwestern University Auditory Test No. 6 (NU-6) in quiet was ≥90% correct by 50% of the participants, with an additional 20% performing at ≥80% correct; the RE performed 1-3% better than the LE. Of the 3291 who completed the WIN on both ears, only 7% exhibited normal performance (50% correct point of ≤6 dB SNR). Overall, WIN performance was significantly better in the RE (mean = 13.3 dB SNR) than in the LE (mean = 13.8 dB SNR). Recognition performance on both the NU-6 and the WIN decreased as a function of both pure-tone hearing loss and age. There was a stronger relation between the high-frequency pure-tone average (1000, 2000, and 4000 Hz) and the WIN than between the pure-tone average (500, 1000, and 2000 Hz) and the WIN. Conclusions: The results on the WIN from both the previous laboratory studies and the current clinical study indicate that the WIN is an appropriate clinic instrument to assess word-recognition performance in background noise. Recognition performance on a speech-in-quiet task does not predict performance on a speech-in-noise task, as the two tasks reflect different domains of auditory function. Experience with the WIN indicates that word-in-noise tasks should be considered the "stress test" for auditory function.

Wireless communications have become an essential part of our daily life. Despite the fact that wireless networks are simple to set up and deploy, the channel conditionsin wireless networks are susceptible to fading and attenuation, thereby reducing the transmission efficiency and reliability. The influence of unstable wireless channels, fading and attenuation is a significant limitation of wireless networks. The persistent and exponential increase in the usage of wireless communication services demands for enhanced reliability, transmission range and efficiency. This can be achieved byexploring and proposing different propagation techniques and models. This thesis evaluates a network model which makes use of an adaptive modulation coding scheme in the presence of an Amplify-and-Forward relaying environment. We deploy adaptive modulation technique in combination with Amplify-and-Forward relaying transmission mechanism to select the best suitable transmission path and obtain better transmission efficiency for a wireless communication system. The network model thus designed comprises of two links, one considering the relay transmission path that travels from the source-to-relay-to-destination, and the other link considering the direct transmission path traveling from the source to the destination, while both the links undergo Nakagami-m fading. In addition, the system ensures better performance in different conditions by making use of adaptive modulation and coding scheme and deploying distinct modulation and coding schemes based on the condition of the communication channel. The performance of this system model is analysed through mathematical analysis and the results are validated and depicted through simulation in MATLAB under different conditions. We have thereby derived the closed-form expressions for the Signal to Noise Ratio (SNR), Outage Probability, and the Packet Error Rate (PER) for the considered network model in the Nakagami-m fading environment. Different fading parameters that affect the system performance are considered and varied while obtaining the required results through MATLAB simulations. The system model proposed significantly reduces the outage probability and packet error rate and aids in achieving better system performance and reliability.

This dissertation presents research on the fabrication of optofluidic sensor platforms, which consist of integrated hollow waveguides and solid waveguides. Antiresonant reflecting optical waveguides (ARROWs) filled with liquids or gases, can be used for high-sensitivity sensing in applications of biotechnology, chemical synthesis, and analytical chemistry. The fabrication method developed for integrated ARROW sensing platforms utilizes standard microfabrication processes and materials. Dielectric cladding layers are deposited on a silicon wafer using plasma-enhanced chemical vapor deposition (PECVD) or sputtering. A sacrificial material is then patterned over the bottom cladding layers by photolithography. Additional dielectric layers are deposited around the core, forming the structure of the waveguides. Integrated solid-core waveguides can be easily created by etching a ridge into the topmost dielectric cladding layer. The hollow core waveguides are then formed by wet etching the sacrificial core material. The coupling efficiency between solid core and hollow core waveguides is extremely important for the platform's overall sensitivity. Efficiencies can be enhanced from 18% to 67% by adjusting the thickness of the thick top oxide. Experimental results prove that optical throughput was improved by 17.1× with this improved interface transmission. Sputtered films were investigated as an alternative to for producing cladding layers. The experimental results reveal that sputtered layers show poor adhesion and mechanical strength which make them unreliable for hollow waveguides with small dimension. High-sensitivity ARROW platforms were obtained by employing hybrid layers (PECVD SiO2 and sputtered Ta2O5) as claddings and building waveguides on self-aligned pedestals. The photolumiscence background was only 1/10 that of previous devices made with SiO2/SiN and the average signal-to-noise ratio was improved by 12×.

In this dissertation, we present analytical and experimental investigations of the electrically actuated micro resonators, when using multi-frequency and/or multi-mode excitation, combined with partial electrodes. We aim to understand their interesting frequency performance and use it to improve the sensing and actuation in microelectromechanical systems (MEMS) and explore their potential applications, such as amplification, gas sensing, magnetometer, multi-physical sensors, and digital-to-analog converters. In the first part, we propose a method of the multi-mode excitation (MME). The concept of the multi-mode excitation is demonstrated by utilizing the superposition principle of two vibration modes in the same phase. To fully understand the difference between the single source excitation (SSE) and multi-mode excitation, we derive the dynamic equations of motions of the electrically-actuated micro cantilever beam and clamped-clamped beam actuated by single/multi-mode excitation. Then, we analytically solve the equations based on the procedure of the Galerkin method with five modes. The simulated results indicate that the MME is clearly superior to that of the SSE, as it can amplify the amplitude performance and signal-to-noise ratio of micro resonators. In the second part, we aim to experimentally prove the concept of the multi-mode excitation and explore its use for gas sensing applications. First, we experimentally investigate the performance of MEMS resonators by single source excitation and multi-mode excitation. We prove the feasibility of the MME approach in enhancing the higher-order mode response for both cantilever and clamped-clamped beams, respectively. We prove that the multi-mode excitation approach provides a better way to activate the higher-order modes with an improved amplitude under a small actuation compared to using a single-source excitation. We then show an improved performance for gas detection. In the third and fourth parts, we propose a technique based on multi-mode excitation for simultaneous sensing for two physical parameters: magnetic field and gas concentration. We respectively investigate a single out-of-plane/in-plane device for in-plane/out-of-plane magnetic field and gas concentration sensing based on tracking the first two vibration modes of a heated buckled micro-beam. We found that operating the resonator at the post-buckling regime, the magnetometer is gas-independent since the first antisymmetric mode (f2) is unaffected by the thermal axial load. Based on it, we utilized the first resonance frequency f1 to detect the gas based on the cooling/heating effects while the second resonance frequency f2 to sense the in-plane/out-of-plane magnetic field. The obtained results demonstrated the sensor acts as a magnetometer and gas sensor, showing good sensitivity, linearity and repeatability. Thus, this technique provides a good candidate for multi-environment monitoring applications. In the last part, we aim to investigate the effects of partial electrodes actuation on the micro resonator and explore its application on the digital-to-analog converter. We analytically and experimentally present modeling, investigation, validation, and optimization of the MEMS resonator-based 3-bit digital to analog converter (DAC) consisting of an in-plane clamped-clamped beam actuated by partial electrodes with different air gaps. The results suggest that the proposed modeling, simulations, and optimization analysis could be successfully implemented in the design of the DAC under various digital combinations. The rich nonlinear behavior with low energy consumption could provide some high potential applications in IoT, such as logic, computation, sensing, and actuation.

No / A proposed method for cancelling or reducing the effect of the transmitted power within the transceiver MIMO unit is investigated and discussed using a feedback signal process between the elements. Several parameters including the level of feedback power mismatch, the transmitted power and the received power are considered in this work to test the performance of the system in term of the bit error rate (BER) versus signal to noise ratio (SNR). A new software programme using MATLAB is implemented to evaluate the proposed method. The results showed that the performances of the system are heavily dependent on the amount of the transmitted power, the received power, and the mismatch in the feedback component.

Although the finite impulse response (FIR) Adaptive Line Enhancer (ALE) was developed in 1975 and has been used in a host of applications, no comprehensive performance analysis has been performed for this method, meaning no general equation exists for its signal-to-noise ratio (SNR) gain. Such an equation would provide practitioners an avenue for determining the amount of noise reduction the ALE provides for a particular application and would add to the general knowledge of adaptive filtering. Based on this motivation, this work derives the general equation for the FIR ALE SNR gain and verifies the equation through computer simulation, under the following assumptions: (1) A simplified Least Mean Squares (LMS) method is used for updating the embedded adaptive filter located within the ALE, (2) The received signal (i.e. the input signal to the ALE) is a summation of sinusoids buried in additive zero-mean white-Gaussian noise (AWGN), (3) The received signal is oversampled (i.e. the sampling rate is larger than the Nyquist rate), and (4) The ALE filter length is an integer multiple of the number of samples within one fundamental period of the original, noiseless signal.

No description available.