Application of improved convolution neural network in financial forecasting

Financial status and its role in the national economy have been increasingly recognized. In order to deduce the source of monetary funds and determine their whereabouts, financial information and prediction have become a scientific method that can not be ignored in the development of national economy. This paper improves the existing CNN and applies it to financial credit from different perspectives. Firstly, the noise of the collected data set is deleted, and then the clustering result is more stable by principal component analysis. The observation vectors are segmented to obtain a set of observation vectors corresponding to each hidden state. Based on the output of PCA algorithm, we recalculate the mean and variance of all kinds of observation vectors, and use the new mean and covariance matrix as credit financial credit, and then determine the best model parameters.The empirical results based on specific data from China's stock market show that the improved convolutional neural network proposed in this paper has advantages and the prediction accuracy reaches.

A method for partitioning trends in genetic mean and variance to understand breeding practices

Quantifying the sources of genetic change is essential for optimising breeding programmes. However, breeding programmes are often complex because many breeding groups are subject to different breeding actions. Understanding the contribution of these groups to changes in genetic mean and variance is essential to understanding genetic change in breeding programmes. Here we extend the previously developed method for analysing the contribution of groups to changes in genetic mean to analysing changes in genetic variance. We, expectedly, show that the contribution of females and males to change in genetic variance can differ and are not independent, indicating we should not look at the contributions in isolation.

Pattern-based downscaling of snowpack variability in the western United States

The decline in snowpack across the western United States is one of the most pressing threats posed by climate change to regional economies and livelihoods. Earth system models are important tools for exploring past and future snowpack variability, yet their coarse spatial resolutions distort local topography and bias spatial patterns of accumulation and ablation. Here, we explore pattern-based statistical downscaling for spatially-continuous interannual snowpack estimates. We find that a few leading patterns capture the majority of snowpack variability across the western US in observations, reanalyses, and free-running simulations. Pattern-based downscaling methods yield accurate, high resolution maps that correct mean and variance biases in domain-wide simulated snowpack. Methods that use large-scale patterns as both predictors and predictands perform better than those that do not and all are superior to an interpolation-based “delta change” approach. These findings suggest that pattern-based methods are appropriate for downscaling interannual snowpack variability and that using physically meaningful large-scale patterns is more important than the details of any particular downscaling method.

Effects of Multi-Shell Free Water Correction on Glioma Characterization

Diffusion MRI is a useful tool to investigate the microstructure of brain tumors. However, the presence of fast diffusing isotropic signals originating from non-restricted edematous fluids, within and surrounding tumors, may obscure estimation of the underlying tissue characteristics, complicating the radiological interpretation and quantitative evaluation of diffusion MRI. A multi-shell regularized free water (FW) elimination model was therefore applied to separate free water from tissue-related diffusion components from the diffusion MRI of 26 treatment-naïve glioma patients. We then investigated the diagnostic value of the derived measures of FW maps as well as FW-corrected tensor-derived maps of fractional anisotropy (FA). Presumed necrotic tumor regions display greater mean and variance of FW content than other parts of the tumor. On average, the area under the receiver operating characteristic (ROC) for the classification of necrotic and enhancing tumor volumes increased by 5% in corrected data compared to non-corrected data. FW elimination shifts the FA distribution in non-enhancing tumor parts toward higher values and significantly increases its entropy (p ≤ 0.003), whereas skewness is decreased (p ≤ 0.004). Kurtosis is significantly decreased (p < 0.001) in high-grade tumors. In conclusion, eliminating FW contributions improved quantitative estimations of FA, which helps to disentangle the cancer heterogeneity.

Psychometric Properties of the General Health Questionnaire (GHQ-28) in Ecuadorian College Students

Objective: To analyse the internal structure of the 28-item version of the General Health Questionnaire (GHQ-28), as well as its reliability and validity in relation to other variables in a sample of Ecuadorian university students. Method: Instrumental design with confirmatory factor analysis using weighted least square mean and variance adjusted (WLSMV) estimator, reliability and convergence and discrimination validity of the GHQ-28. Sample: 495 students (56.6% women), between 18 to 35 years old (M = 24.1 years; SD = 2.1), from three universities (59.6% public) in Ecuador. Results: The bifactor model of the GHQ-28 test has an adequate fit with χ2 = 357.81; p &gt; .05; df = 322; χ2/df = 1.11; CFI = .991; TLI = .989; SRMR = .059; RMSEA = .015 [.000 – .023]; ωH = .93; ECV = .90; PUC = .78. The GHQ-28 is reliable and in terms of convergent validity, it correlates significantly and negatively with mental health, assessed by MHC-SF, and it is discriminant between risk and non-risk cases. Conclusion: The GHQ-28 bifactor model is replicable in Ecuadorian college students.

RELATIVE RISK OF CORONAVIRUS DISEASE (COVID-19) IN SOUTH SULAWESI PROVINCE, INDONESIA: BAYESIAN SPATIAL MODELING

The Covid-19 has exploded in the world since late 2019. South Sulawesi Province has the highest number of Covid-19 cases outside Java Island in Indonesia. This paper aims to determine the most suitable Bayesian spatial conditional autoregressive (CAR) localised models in modeling the relative risk (RR) of Covid-19 in South Sulawesi Province, Indonesia. Bayesian spatial CAR localised models with different hyperpriors were performed adopting a Poisson distribution for the confirmed Covid-19 counts to examine the grouping of Covid-19 cases. All confirmed cases of Covid-19 (19 March 2020-18 February 2021) for each district were included. Overall, Bayesian CAR localised model with G = 5 with a hyperprior IG (1, 0.1) is the preferred model to estimate the RR based on the two criteria used. Makassar and Toraja Utara have the highest and the lowest RR, respectively. The group formed in the localised model is influenced by the magnitude of the mean and variance in the count data between areas. Using suitable Bayesian spatial CAR localised models enables the identification of high-risk areas of Covid-19 cases. This localised model could be applied in other case studies.

Statistical Model of Correlation Difference and Related-Key Linear Cryptanalysis

The goal of this work is to propose a related-key model for linear cryptanalysis. We start by giving the mean and variance of the difference of sampled correlations of two Boolean functions when using the same sample of inputs to compute both correlations. This result is further extended to determine the mean and variance of the difference of correlations of a pair of Boolean functions taken over a random data sample of fixed size and over a random pair of Boolean functions. We use the properties of the multinomial distribution to achieve these results without independence assumptions. Using multivariate normal approximation of the multinomial distribution we obtain that the distribution of the difference of related-key correlations is approximately normal. This result is then applied to existing related-key cryptanalyses. We obtain more accurate right-key and wrong-key distributions and remove artificial assumptions about independence of sampled correlations. We extend this study to using multiple linear approximations and propose a Χ2-type statistic, which is proven to be Χ2 distributed if the linear approximations are independent. We further examine this statistic for multidimensional linear approximation and discuss why removing the assumption about independence of linear approximations does not work in the related-key setting the same way as in the single-key setting.

Mean and variance of the Eulerian and Lagrangian horizontal velocities induced by nonlinear multi-directional irregular water waves

In this paper, we study the mean and variance of the Eulerian and Lagrangian fluid velocities as a function of depth below the surface of directionally spread irregular wave fields given by JONSWAP spectra in deep water. We focus on the behaviour of these quantities in the bulk of the water, and using second-order potential flow theory we derive new simple asymptotic approximations for their decay in the limit of large depth below the surface. Specifically, we show that when the depth is greater than about 1.5 peak wavelengths, the variance of the Eulerian velocity decays in proportion to exp ⁡ ( − ( 135 4 ) 1 / 3 ( − k p z ) 2 / 3 ) , and the mean Lagrangian velocity decays in proportion to 1 ( − k p z ) 1 / 6 exp ⁡ ( − ( 135 4 ) 1 / 3 ( − k p z ) 2 / 3 ) . Here, k p is the peak wave number and z is the vertical coordinate measured positively upwards from the still water level. We test the accuracy of the second-order formulation against new fully nonlinear simulations of both short crested and long crested irregular wave fields and find a good match, even when the simulations are known to be affected substantially by third-order effects. To our knowledge, this marks the first fully nonlinear investigation of the Eulerian and Lagrangian velocities below the surface in irregular wave fields.

Effortless integration of probabilistic visual input

Prominent theories of perception suggest that the brain builds probabilistic models of the world, assessing the statistics of the visual input to inform this construction. However, the evidence for this idea is often based on simple impoverished stimuli, and the results have often been discarded as an illusion reflecting simple "summary statistics" of visual inputs. Here we show that the visual system represents probabilistic distributions of complex heterogeneous stimuli. Importantly, we show how these statistical representations are integrated with representations of other features and bound to locations, and can therefore serve as building blocks for object and scene processing. We uncover the organization of these representations at different spatial scales by showing how expectations for incoming features are biased by neighboring locations. We also show that there is not only a bias, but also a skew in the representations, arguing against accounts positing that probabilistic representations are discarded in favor of simplified summary statistics (e.g., mean and variance). In sum, our results reveal detailed probabilistic encoding of stimulus distributions, representations that are bound with other features and to particular locations.

Comparison between methods for measuring fecal egg count and estimating genetic parameters for gastrointestinal parasite resistance traits in sheep

Fecal egg count (FEC) is an indicative measurement for parasite infection in sheep. Different FEC methods may show inconsistent results. Not accounting for inconsistencies can be problematic when integrating measurements from different FEC methods for genetic evaluation. The objectives of this study were to evaluate the difference in means and variances between two fecal egg counting methods used in sheep, the Modified McMaster (LMMR) and the Triple Chamber McMaster (LTCM); to estimate variance components for the two FEC methods, treating them as two different traits; and to integrate FEC data from the two different methods and estimate genetic parameters for FEC and other gastrointestinal parasite resistance traits. Fecal samples were collected from a commercial Rideau-Arcott sheep farm in Ontario. Fecal egg counting was performed using both Modified McMaster and the Triple Chamber McMaster methods. Other parasite resistance trait records were collected from the same farm including eye score (FAMACHA ©), body condition score (BCS), and body weight (WT). The two FEC methods were highly genetically (0.94) and phenotypically (0.88) correlated. However, the mean and variance between the two FEC methods were significantly different (P &lt; 0.0001). Therefore, re-scaling is required prior to integrating data from the different methods. For the multiple trait analysis, data from the two fecal egg counting methods were integrated (LFEC) by using records for the LMMR when available and replacing missing records with re-standardized LTCM records converted to the same mean and variance of LMMR. Heritability estimates were 0.12 ± 0.04, 0.07 ± 0.05 , 0.17 ± 0.06, and 0.24 ± 0.07 for LFEC egg count, FAMACHA ©, BCS, and WT, respectively. The estimated genetic correlations between fecal egg count and the other parasite resistance traits were low and not significant (P&gt;0.05) for FAMACHA © (r= 0.24 ± 0.32) and WT (r= 0.22 ± 0.19), and essentially zero for BCS (r= -0.03 ± 0.25), suggesting little to no benefit of using such traits as indicators for LFEC.