Log-likelihood Ratio Method Research Articles

Incidence and determinants of diabetic ketoacidosis among people with diabetes in Woldiya comprehensive specialized hospital, Ethiopia: a retrospective cohort study

IntroductionDiabetic ketoacidosis is an acute complication of diabetes mellitus that is characterised by hyperglycemia, acidosis, and ketonuria. Diabetes is the most challenging public health problem in the twenty-first century for both developed and developing countries.ObjectiveTo assess the incidence of Diabetic ketoacidosis and its determinants among adult people with diabetes at an Ethiopian Hospital.MethodAn institution-based retrospective cohort study was conducted among 390 adult people with diabetes attending services at Wolida Comprehensive Specialized Hospital. The consecutive sampling method was used to select study participant charts. Data were collected using a checklist prepared from different literature. The data were entered into EPI data version 4.6.0.5 and exported to STATA version 14.0 for further analysis. The Wiebull model was the best fitted model that was selected using the log-likelihood ratio method and the Akakian information criterion. Hazard ratios with their 95% confidence interval and p-value were computed.ResultFrom the total 405 charts reviewed, 390 adult charts were included for analysis. A total of 121 DKA occurred from 5471 person–months of observation. The overall incidence rate of diabetic ketoacidosis was found to be 2.2 per 100 person-months (95% CI: 1.8- 2.6). Being urban dweller (AHR: 0.59, 95% CI: 0.36–0.99), having no family history of DM (AHR: 0.55, 95%CI: 0.31—0.97), presence of infection (AHR: 2.60, 95%CI = 1.06–6.39), having of any comorbidities (AHR: 4.31, 95% CI: 1.70–10.90), and having poor glycemic control (AHR: 7.45, 95% CI: 3.84–14.47) were significant determinants.Conclusion and recommendationsThe overall incidence of diabetic ketoacidosis in study area was relatively high. Poor glycemic control, the presence of infection, and comorbidity were determinants of diabetic ketoacidosis. There is a need to have close follow-up of people with diabetes who have comorbidity, infection, and poor glycemic control.

Chinese technical terminology extraction based on DC-value and information entropy

China's technology is developing rapidly, and the number of patent applications has surged. Therefore, there is an urgent need for technical managers and researchers that how to apply computer technology to conduct in-depth mining and analysis of lots of Chinese patent documents to efficiently use patent information, perform technological innovation and avoid R&D risks. Automatic term extraction is the basis of patent mining and analysis, but many existing approaches focus on extracting domain terms in English, which are difficult to extend to Chinese due to the distinctions between Chinese and English languages. At the same time, some common Chinese technical terminology extraction methods focus on the high-frequency characteristics, while technical domain correlation characteristic and the unithood feature of terminology are given less attention. Aiming at these problems, this paper proposes a Chinese technical terminology method based on DC-value and information entropy to achieve automatic extraction of technical terminology in Chinese patents. The empirical results show that the presented algorithm can effectively extract the technical terminology in Chinese patent literatures and has a better performance than the C-value method, the log-likelihood ratio method and the mutual information method, which has theoretical significance and practical application value.

A Study of Cognitive Radio Sensing Techniques for Optimum Spectrum Utilization

Background: The spectrum scarcity plays a vital role in wireless communications. We are in the situation to use it through an efficient methodology. Objective: To identify the holes in the spectrum through an efficient spectrum sensing and to allocate the bands to the unlicensed users (Secondary Users- SU). Methods: It has been proposed to make a comparative study among the existing spectrum sensing methods based on the following parameters such as Probability detection (Pd) measurement, Algorithm, Decision fusion method, Network model. Results: Comparative study has been made to find the pros and cons of the existing techniques with their limitations and the field of application. CSS (Cooperative Spectrum Sensing) technique significantly consumes less energy and reporting time to FC (Fusion Centre) since it utilizes LLR (Log Likelihood Ratio) Method to find the Probability of detection using Chair-varshney rule in local sensing with parallel report approach in the cluster based network. Conclusion: Through the study and the comparison of parameters in literature, it is found that CSS provide better detection. Therefore, this technique can be considered as an efficient technique to find the holes and to share the frequency with SU’s.

Sensor fault diagnosis for lithium-ion battery packs based on thermal and electrical models

Sensor fault diagnosis is a crucial technology for the battery management system. In this work, a sensor fault diagnosis scheme is proposed for the battery pack using equivalent models and particle filters. The thermal and electrical models of the battery pack are developed based on the Thevenin equivalent circuit model and the radial equivalent thermal model. The model parameters are identified by the recursive least square algorithm. The particle filter estimates the temperature and voltage of the battery pack, which overcomes the problems of system noise and nonlinearity. The studentized residual method based on a sliding window is developed to eliminate the residual outliers caused by uncertainty. Monte Carlo simulation is used to calculate the specific values of distribution function, mean and standard deviation for different sequences. The critical value table of the absolute value method of studentized residuals is obtained by interpolating. The cumulative sum of residual log-likelihood ratio method based on sliding windows is developed to calculate residuals. By multiple residual assessments, faults of voltage, current, temperature sensors, and batteries are detected and isolated. Different fault cases are simulated to verify the proposed scheme. The results of the experiment and simulation results verify the effectiveness of the proposed sensor fault diagnosis scheme.

Small sample inference for the common coefficient of variation

This paper utilizes the modified signed log-likelihood ratio method for the problem of inference about the common coefficient of variation in several independent normal populations. This method is applicable for both the problem of hypothesis testing and constructing a confidence interval for this parameter. Simulation studies show that the coverage probability of this proposed approach is close to the confidence coefficient. Also, its expected length is smaller than expected lengths of other competing approaches. In fact, the proposed approach is very satisfactory regardless of the number of populations and the different values of the common coefficient of variation even for very small sample size. Finally, we illustrate the proposed method using two real data sets.

Enhanced Coordinate Interleaved OFDM With Index Modulation

Orthogonal frequency-division multiplexing with index modulation (OFDM-IM) has attracted much attention in recent years. In OFDM-IM, the information bits of a block are generally divided into several subblocks, each of which is further split into two parts, i.e., the index bits and the symbol bits. By performing a coordinate interleaving operation over each two constellation symbols, the bit-error-rate (BER) performance of the symbol bits is significantly improved in an existing OFDM-IM scheme, which is called coordinate-interleaved OFDM-IM (CI-OFDM-IM). In this paper, we propose a novel OFDM-IM scheme termed enhanced CI-OFDM-IM (ECI-OFDM-IM). In ECI-OFDM-IM, to improve the BER performance of the index bits, we adopt a special subcarrier-activation pattern (SAP) lookup table in which the Hamming distance of any pair of SAPs is at least four. To improve the BER performance of the symbol bits conveyed by the active subcarriers, the operations of rotating and coordinate interleaving are conducted twice over each four constellation symbols. Based on the maximum likelihood (ML) detection, a theoretical upper bound on the average bit-error probability of ECI-OFDM-IM is derived in closed form. By using the log-likelihood ratio method for SAP detection, a semi-ML detection scheme is proposed. Both the theoretical and simulation results show that ECI-OFDM-IM outperforms CI-OFDM-IM with additional diversity gains over frequency-selective Rayleigh fading channels.

On Blind Recognition of Channel Codes Within a Candidate Set

Existing schemes for blind recognition of channel codes make use of the average log-likelihood ratio (LLR) of each code’s parity checks. There are difficulties in setting necessary thresholds and in theoretical analysis of the method. This letter proposes to use the average likelihood difference (LD) of the parity checks. The computational complexity is reduced while the recognition performance is kept comparable to that of the LLR method. Moreover, an optimum decoding-based blind recognition is also described. Simulations show that, for rate 1/2 convolutional codes, this new method notably improves the performance.

Parzen window entropy based spectrum sensing in cognitive radio

In this paper, we propose a Parzen window entropy based spectrum sensing algorithm for enhancing the signal-to-noise ratio (SNR) wall of cognitive radio primary user detection. We compute the information entropy using a non-parametric Kernel Density Estimation (KDE) method. Single node sensing is extended to cooperative sensing using the weighted gain combining (WGC) fusion method. The weights of WGC technique are computed using a Differential Evolution(DE) algorithm and compared with the log-likelihood ratio (LLR) method. In addition, the detection performance of the proposed Parzen window entropy is compared with Shannon entropy and energy detection techniques. We consider a DVB-T signal with Additive White Gaussian Noise (AWGN) subjected to Rayleigh fading under noise uncertainty as a primary user signal for simulation. The simulation result reveals that in the case of a single node and cooperative sensing, the proposed method achieves SNR wall of −19 dB and −24 dB respectively at the probability of false alarm 0.1.

Decision directed iterative channel estimation and Reed–Solomon decoding over flat fading channels

Iterative channel estimation and decoding over flat Rayleigh fading channels have been shown in literature to improve the performance of receivers in wireless communication systems. Turbo and low density parity check codes have mostly been used as the forward error correction (FEC) scheme in the iterative receiver structures. This study proposes a decision directed iterative channel estimation and decoding receiver using Reed–Solomon codes. In particular, the Koetter and Vardy, Reed–Solomon (KV–RS) soft decision decoder is adopted in the receiver structure as the FEC scheme. Two methods of deriving a priori information in joint iterative channel estimation and decoding receiver structures are also analysed in this study. The first method derives the a priori information employing the log likelihood ratio (LLR) method. A priori information derived using the LLR method is mostly used in iterative receiver structures in the literature. The second method derives the a priori information employing the distance metric (DM) method. The DM method is proposed in this study as an alternative approach to the LLR method of deriving a priori information in the iterative receivers. Hard and soft feedback information from the KV–RS decoder is considered to verify the performance of these methods. The performance of the two a priori information methods is documented through computer simulation assuming an M-ary quadrature amplitude modulation system. Simulation results verify the improvement in symbol error rate (SER) performance in the KV–RS iterative receiver structure in comparison with the KV–RS receiver without feedback. Also, the proposed DM method exhibits the same SER performance in comparison with the LLR method which is mostly used in the literature. More importantly, the proposed DM method has the advantage of less computational delay and time complexity compared with the LLR method.

Confidence Intervals of Effect Size for Randomized Comparative Parallel-Group Studies with Unequal Variances

Recent years have seen a heightened interest in estimating effect size—a common measure of effect magnitude in biomedical research—because of its direct clinical relevance. In this article, three interval estimates of effect size for randomized comparative parallel-group studies with unequal variances are discussed. Two real-life examples illustrate that confidence intervals obtained by three methods are quite different, especially when the sample sizes are small. Simulation results show that confidence intervals generated by the modified signed log-likelihood ratio method yield essentially the exact coverage probabilities, whereas the other two methods, even though they are more popular methods, yield less satisfactory results.

Inference for the Normal Mean with Known Coefficient of Variation

Inference for the mean of a normal distribution with known coefficient of variation is of special theoretical interest be- cause the model belongs to the curved exponential family with a scalar parameter of interest and a two-dimensional minimal sufficient statistic. Therefore, standard inferential methods cannot be directly applied to this problem. It is also of practical interest because this problem arises naturally in many environmental and agriculture studies. In this paper we proposed a modified signed log likelihood ratio method to obtain inference for the normal mean with known coeffi- cient of variation. Simulation studies show the remarkable accuracy of the proposed method even for sample size as small as 2. Moreover, a new point estimator for the mean can be derived from the proposed method. Simulation studies show that new point estimator is more efficient than most of the existing estimators.

Psychological distress across the lifespan: examining age-related item bias in the Kessler 6 Psychological Distress Scale

Old age respondents may differ systemically in their responses to measures of psychological distress over and above their actual latent distress levels when compared to younger respondents. The current study aimed to investigate the potential for age-related bias(es) in the Kessler 6 Psychological Distress Scale (K6) items. Data from the 2007 Australian National Survey of Mental Health and Wellbeing were analyzed using Item Response Theory to detect the presence of item bias in each of the K6 items. The potential for item bias was assessed by systematically comparing respondents classed as young (16-34 years), middle aged (35-64 years), and old aged (65-85 years). The significance and magnitude of the item bias between the age groups was assessed using the log-likelihood ratio method of differential item functioning. After statistical adjustment, there were no biases of significant magnitude influencing the endorsement of K6 items between young and middle-aged respondents or between middle-aged and old age respondents. There was a bias of significant magnitude present in the endorsement of the K6 item addressing levels of fatigue between young and old age respondents. Despite the identification of significant item bias in the endorsement of K6 items between the age groups, the magnitude and influence of the bias on total K6 scores is likely to have little influence on the overall interpretation of group data when comparing psychological distress across the lifespan. Researchers should be cautious, however, when examining individual levels of fatigue related to psychological distress in older individuals.

Confidence Intervals of Effect Size for Paired Studies

To construct a confidence interval of effect size in paired studies, we propose four approximate methods--Wald method, variance-stabilizing transformation method, and signed and modified signed log-likelihood ratio methods. We compare these methods using simulation to determine those that have good performance in terms of coverage probability. In particular, simulations show that the modified signed log-likelihood ratio method produces a confidence interval with a nearly exact coverage probability and highly accurate and symmetric error probabilities even for very small samples. We apply the methods to data from an iron deficiency anemia study.

Evaluating measurement equivalence using the item response theory log-likelihood ratio (IRTLR) method to assess differential item functioning (DIF): applications (with illustrations) to measures of physical functioning ability and general distress

Methods based on item response theory (IRT) that can be used to examine differential item functioning (DIF) are illustrated. An IRT-based approach to the detection of DIF was applied to physical function and general distress item sets. DIF was examined with respect to gender, age and race. The method used for DIF detection was the item response theory log-likelihood ratio (IRTLR) approach. DIF magnitude was measured using the differences in the expected item scores, expressed as the unsigned probability differences, and calculated using the non-compensatory DIF index (NCDIF). Finally, impact was assessed using expected scale scores, expressed as group differences in the total test (measure) response functions. The example for the illustration of the methods came from a study of 1,714 patients with cancer or HIV/AIDS. The measure contained 23 items measuring physical functioning ability and 15 items addressing general distress, scored in the positive direction. The substantive findings were of relatively small magnitude DIF. In total, six items showed relatively larger magnitude (expected item score differences greater than the cutoff) of DIF with respect to physical function across the three comparisons: "trouble with a long walk" (race), "vigorous activities" (race, age), "bending, kneeling stooping" (age), "lifting or carrying groceries" (race), "limited in hobbies, leisure" (age), "lack of energy" (race). None of the general distress items evidenced high magnitude DIF; although "worrying about dying" showed some DIF with respect to both age and race, after adjustment. The fact that many physical function items showed DIF with respect to age, even after adjustment for multiple comparisons, indicates that the instrument may be performing differently for these groups. While the magnitude and impact of DIF at the item and scale level was minimal, caution should be exercised in the use of subsets of these items, as might occur with selection for clinical decisions or computerized adaptive testing. The issues of selection of anchor items, and of criteria for DIF detection, including the integration of significance and magnitude measures remain as issues requiring investigation. Further research is needed regarding the criteria and guidelines appropriate for DIF detection in the context of health-related items.

Confidence intervals of effect size in randomized comparative parallel‐group studies

We have presented a new likelihood-based approach for constructing confidence intervals of effect size that are applicable to small samples. We also conduct a simulation study to compare the coverage probability of the new likelihood-based method with other three methods proposed by Hedges and Olkin and Kraemer and Paik. Simulation studies show that the confidence interval generated by the modified signed log-likelihood ratio method possesses essentially exact coverage probabilities even for small samples, although the coverage probabilities are consistently but slightly less than the nominal level. The methods are also applied to two examples.

Likelihood-based confidence intervals for a log-normal mean.

To construct a confidence interval for the mean of a log-normal distribution in small samples, we propose likelihood-based approaches - the signed log-likelihood ratio and modified signed log-likelihood ratio methods. Extensive Monte Carlo simulation results show the advantages of the modified signed log-likelihood ratio method over the signed log-likelihood ratio method and other methods. In particular, the modified signed log-likelihood ratio method produces a confidence interval with a nearly exact coverage probability and highly accurate and symmetric error probabilities even for extremely small sample sizes. We then apply the methods to two sets of real-life data.

Cerebral ventricular enlargement as a generalized feature of schizophrenia: a distribution analysis on 502 subjects

Enlargement of cerebral ventricles is one of the most replicated biological features, and the one quantitatively most deviant in schizophrenia. It occurs in the early phases of the disease and may have pathogenetic relevance. Whether this abnormality is limited to a specific subgroup of patients or is a common feature to most or all patients affected by schizophrenia, however, is still a matter of debate. The answer to this question would improve our comprehension of the nature of this abnormality and contribute to the debate between the competing hypotheses of biological homogeneity vs heterogeneity of schizophrenia. We performed a distribution analysis of lateral ventricular dimensions of 340 schizophrenic patients and 162 non-psychiatric controls. All subjects underwent cerebral computerized tomographic scan, and ventricular dimensions were expressed as ventricular brain ratio (VBR). After removing the effect of confounding variables (age, sex and type of scanner) on individual VBR, data were power-transformed and different distribution hypotheses were tested by means of the maximum log-likelihood ratio method. Our findings indicate that, in the mixed sample of patients and controls, a mixture of two gaussian curves represents the distribution better than a single gaussian curve, but no evidence emerged leading to rejection of the normality hypothesis in the schizophrenic patients sample. Lateral ventricular enlargement in schizophrenia is not a marker of a discrete subgroup of schizophrenia, but occurs in most, if not all, schizophrenic patients. This supports the hypothesis of biological homogeneity of the disease, at least relative to its major brain morphological abnormality.

Intertemporal Changes in the Riskiness of REITs

This study investigates the variability in the risk components of REITs over the 1973-1989 period using the cusum test, the cusum of squares test, and the Quandt's log-likelihood ratio method. Four...

Likelihood ratios as a measure of the diagnostic usefulness of excretory urogram information.

The log likelihood ratio (LLR) method of assessing physicians' diagnostic judgments can be used to measure the diagnostic usefulness of radiologic information (urography in this study). The method requires collection of referring pbysicians' diagnosis prior to and after urography and their certainty in relation to receipt of the radiologic information. Physicians in daily practice are able to provide diagnoses and assign a percentage number to their certainty. The LLR method permits measurement of the diagnostic usefulness of both normal and abnormal urogram results. Retrospective chart review does not provide the same LLR results as those obtained from the prospective study.