scholarly journals Timeseries Forecasting using Long Short-Term Memory Optimized by Multi Heuristics Algorithm

2019 ◽  
Vol 8 (4) ◽  
pp. 11492-11500
Keyword(s):  
Deep Learning ◽  
Mutual Funds ◽  
Mutual Fund ◽  
Short Term Memory ◽  
Future Price ◽  
Optimization Approach ◽  
Optimal Learning ◽  
Asset Value ◽  
Heuristics Algorithm ◽  
Lstm Network

Forecasting future price of financial instruments (such as equity, bonds and mutual funds) has become an ongoing effort of financial and capital market industry members. The most current technology is usually applied by high economic scale companies to solve the ambitious and complicated problem. This paper presents optimization solution for a deep learning model in forecasting selected Indonesian mutual funds' Net Asset Value (NAV). There is a well-known issue in determining a deep learning parameters in LSTM network like window timestep and number of neurons to be used in getting the optimal learning from the historical data. This research tries to provide solution by utilizing multi-heuristics optimization approach consists of Genetic Algorithm (GA) and Particle Swarm Optimization (PSO) to determine the best LSTM's network parameters, namely window timesteps and number of neurons. The result shows that from the nine selected mutual funds, PSO outperforms GA in optimizing the LSTM model by giving a lower Root Square Mean Error (RMSE) by 460.84% compared to GA's. However, PSO took a longer execution time by 1.78 times of GA's. This paper also confirms that based on RMSE for both training and evaluation dataset, equity mutual fund's forecasted NAV has the highest RMSE followed by fixed income mutual fund's forecasted NAV and money market mutual fund forecasted NAV.

scholarly journals Real-Time Detection of Dictionary DGA Network Traffic Using Deep Learning

2021 ◽  
Vol 2 (2) ◽  
Author(s):  
Kate Highnam ◽  
Domenic Puzio ◽  
Song Luo ◽  
Nicholas R. Jennings
Keyword(s):  
Neural Network ◽  
Deep Learning ◽  
Real Time ◽  
Network Traffic ◽  
Short Term Memory ◽  
Domain Names ◽  
Control Networks ◽  
Detection Techniques ◽  
Lstm Network ◽  
And Control

AbstractBotnets and malware continue to avoid detection by static rule engines when using domain generation algorithms (DGAs) for callouts to unique, dynamically generated web addresses. Common DGA detection techniques fail to reliably detect DGA variants that combine random dictionary words to create domain names that closely mirror legitimate domains. To combat this, we created a novel hybrid neural network, Bilbo the “bagging” model, that analyses domains and scores the likelihood they are generated by such algorithms and therefore are potentially malicious. Bilbo is the first parallel usage of a convolutional neural network (CNN) and a long short-term memory (LSTM) network for DGA detection. Our unique architecture is found to be the most consistent in performance in terms of AUC, $$F_1$$ F 1 score, and accuracy when generalising across different dictionary DGA classification tasks compared to current state-of-the-art deep learning architectures. We validate using reverse-engineered dictionary DGA domains and detail our real-time implementation strategy for scoring real-world network logs within a large enterprise. In 4 h of actual network traffic, the model discovered at least five potential command-and-control networks that commercial vendor tools did not flag.

scholarly journals Detection of dementia on raw voice recordings using deep learning: A Framingham Heart Study

2021 ◽  
Author(s):  
Chonghua Xue ◽  
Cody Karjadi ◽  
Ioannis Ch. Paschalidis ◽  
Rhoda Au ◽  
Vijaya B. Kolachalama
Keyword(s):  
Deep Learning ◽  
Framingham Heart Study ◽  
Short Term Memory ◽  
Neuropsychological Testing ◽  
Heart Study ◽  
Audio Recordings ◽  
Lstm Network ◽  
Sensitivity Specificity ◽  
Normal Cognition ◽  
Mean Sensitivity

AbstractBackgroundIdentification of reliable, affordable and easy-to-use strategies for detection of dementia are sorely needed. Digital technologies, such as individual voice recordings, offer an attractive modality to assess cognition but methods that could automatically analyze such data without any pre-processing are not readily available.MethodsWe used a subset of 1264 digital voice recordings of neuropsychological examinations administered to participants from the Framingham Heart Study (FHS), a community-based longitudinal observational study. The recordings were 73 minutes in duration, on average, and contained at least two speakers (participant and clinician). Of the total voice recordings, 483 were of participants with normal cognition (NC), 451 recordings were of participants with mild cognitive impairment (MCI), and 330 were of participants with dementia. We developed two deep learning models (a two-level long short-term memory (LSTM) network and a convolutional neural network (CNN)), which used the raw audio recordings to classify if the recording included a participant with only NC or only dementia, and also to differentiate between recordings corresponding to non-demented (NC+MCI) and demented participants.FindingsBased on 5-fold cross-validation, the LSTM model achieved a mean (±std) area under the sensitivity-specificity curve (AUC) of 0.744±0.038, mean accuracy of 0.680±0.032, mean sensitivity of 0.719±0.112, and mean specificity of 0.652±0.089 in predicting cases with dementia from those with normal cognition. The CNN model achieved a mean AUC of 0.805±0.027, mean accuracy of 0.740±0.033, mean sensitivity of 0.735±0.094, and mean specificity of 0.750±0.083 in predicting cases with only dementia from those with only NC. For the task related to classification of demented participants from non-demented ones, the LSTM model achieved a mean AUC of 0.659±0.043, mean accuracy of 0.701±0.057, mean sensitivity of 0.245±0.161 and mean specificity of 0.856±0.105. The CNN model achieved a mean AUC of 0.730±0.039, mean accuracy of 0.735±0.046, mean sensitivity of 0.443±0.113, and mean specificity of 0.840±0.076 in predicting cases with dementia from those who were not demented.InterpretationThis proof-of-concept study demonstrates the potential that raw audio recordings of neuropsychological testing performed on individuals recruited within a community cohort setting can provide a level of screening for dementia.

scholarly journals Evaluation of Mutual Funds Performance And Consistency test over the use of Performance Sizing Methods

2020 ◽  
Vol 5 (6) ◽  
pp. 650-656
Author(s):  
Eka Kusumawati ◽  
Ega Bagja Nugraha
Keyword(s):  
Mutual Funds ◽  
Mutual Fund ◽  
Sharpe Ratio ◽  
Superior Performance ◽  
Ratio Method ◽  
Surveillance Period ◽  
Consistency Test ◽  
Asset Value ◽  
Equity Funds ◽  
Coefficient Of Concordance

The development of mutual fund industry in Indonesia has increases every year. From those several types of equity funds, the Net Asset Value (NAV) of mutual funds has increased by quite high number from year to year compared to other types. This research was assess the performance of mutual funds and examine those several consistency over the use of performance sizing methods from Sharpe ratio, Treynor index and Jensen's Alpha methods. Current problem who was stumbled was how the performance of stock mutual funds was measured by the Sharpe ratio, Treynor index and Jensen's Alpha methods and whether there has consistency over its performance by using it. The recent sample was 37 mutual funds that were registered at BAPEPAM-LK and still operating in Indonesia from January 2009 to October 2013. Performance evaluations used Sharpe ratio method, Treynor index and Jensen's Alpha. As for assess those consistency of the use performance sizing methods was done by Kendall coefficient of concordance (W) test. The result over this research said that Panin Dana Maksima and Panin Dana Prima are the best mutual funds, this could be seen during these surveillance period which found that mutual fund has superior performance above the market. The result of consistency test over those performance of stock mutual funds using Kendall W's concordance coefficient found that there has consistency or harmony when evaluated the performance of equity funds by using Sharpe Ratio, Treynor Index and Jensen's Alpha methods during those period.

Pengaruh Inflasi, BI-7 Day Reverse Repo Rate, Kurs, Jumlah Uang Beredar dan Indeks Harga Saham Gabungan terhadap Nilai Aktiva Bersih Reksa Dana Saham

Owner ◽  
2021 ◽  
Vol 5 (2) ◽  
pp. 358-367
Author(s):  
Jhon Lismart Benget. P.
Keyword(s):  
Exchange Rate ◽  
Mutual Funds ◽  
Money Supply ◽  
Mutual Fund ◽  
Stock Price ◽  
Price Index ◽  
Asset Value ◽  
Net Asset Value ◽  
Repo Rate ◽  
Stock Price Index

The purpose of this study is to examine the effect of inflation, BI-7 day reverses repo rate, exchange rate, the money supply, and composite stock price index on the net asset value of stock mutual funds. The population of this study is the stock mutual fund which was listed on the financial services authority in 2017-2020. The results of this study indicate that simultaneously inflation, BI-7 day reverse repo rate, exchange rate, the money supply, and composite stock price index affect the net asset value of the stock mutual fund. Partially, this study show BI-7 day reverse repo rate has a positive and significant effect on the net asset value of a stock mutual fund. The exchange rate has a positive and significant effect on the net asset value of stock mutual funds. The composite stock price index has a positive and significant effect on the net asset value of stock mutual funds. The money supply has a negative and significant effect on the net asset value of a stock mutual fund while inflation has no significant effect on the net asset value of a stock mutual fund.

Indonesian Mutual Funds Classification Using Clustering Method

2015 ◽  
Vol 21 (4) ◽  
pp. 826-829
Author(s):  
Ir. Dewi Tamara ◽  
Shintia Revina
Keyword(s):  
Mutual Funds ◽  
Mutual Fund ◽  
Money Market ◽  
Fixed Income ◽  
Clustering Method ◽  
New Classification ◽  
Asset Value ◽  
Net Asset Value ◽  
Alternative Investment

Mutual funds have existed since 1990 as an alternative investment in Indonesia. The objective of this research is to examine the existing classification of mutual funds database. The data of mutual funds is taken from Bloomberg through Portal Reksadana 2013 which covered 690 mutual funds. The existing classification consists of mutual funds fixed income (reksadana pendapatan tetap), equity (reksadana saham), money market (reksadana pasar uang) and structured (reksadana campuran). The existing financial attributes consists of the net asset value, percentage annualized return the last 6 months, 1 year, 3 years, 5 years and year-to-date. This paper uses K-means clustering to propose new classification of Indonesian mutual funds. The result reveals that mutual funds in equity and fixed income belong to its group. However, mutual funds money market is belong to mutual fund fixed income and mutual funds structures are identified to mutual funds equity. Furthermore, we find that in average 43% of Indonesian mutual funds are misclassified in accordance with their attributes. Finally, it is suggested to re-group the mutual funds into smaller classification, which has lower rates of misclassified mutual funds and possibility to achieve better performances in terms of its percentage annualized return.

scholarly journals Exchange-traded funds as an alternative investment option

2019 ◽  
pp. 7-37
Author(s):  
António Afonso ◽  
Pedro Cardoso
Keyword(s):  
Mutual Funds ◽  
Mutual Fund ◽  
Exchange Traded Funds ◽  
Price Efficiency ◽  
Index Funds ◽  
Equity Mutual Funds ◽  
Asset Value ◽  
Net Asset Value ◽  
Tracking Ability ◽  
Investment Option

We conduct an analysis of Exchange-traded Funds (ETFs), Index and Equity mutual funds and their respective benchmark during the 2010-2015 period for the Portuguese fund industry. For the period 2010-2017, we test ETFs for price inefficiency (existence of deviations between prices and the Net Asset Value) and persistence. We find that the studied ETF does not always outperform index funds in replicating the variations of the PSI 20 index, despite exhibiting better tracking ability when facing downside deviations of the benchmark and a better capacity of smoothing tracking deviations. Regarding ETFs price efficiency and its persistence, the study reveals that the examined ETF is priced at a low average discount with evidence of deviations persistence of at least two days. The investment schemes with the highest ability to track the PSI 20 Index were PSI20 (ETF), BBVA PPA Índice PSI20, and the equity mutual fund BPI Portugal.

scholarly journals Growth and Value Effect on Jakarta Islamic Index: Analysis towards Performance of Sharia Equity Mutual Fund

2019 ◽  
Vol 11 (1) ◽  
Author(s):  
Sylva Alif Rusmita ◽  
Marhanum Che Mohd Salleh
Keyword(s):  
Mutual Funds ◽  
Mutual Fund ◽  
Secondary Data ◽  
Additional Insight ◽  
Risk And Return ◽  
Fund Managers ◽  
Index Analysis ◽  
Asset Value ◽  
Net Asset Value ◽  
Value Effect

This study provides evidence that value and stocks’ growth able to explain Net Asset Value of Shariah Mutual Fund. It is important for investment managers and investors to estimate future profit or loss that may happen on their mutual funds prior they venture into the investment platform. This study therefore is conducted to prove that factors including value and growth may affect the future profit of Shariah Mutual Funds. Based on quantitative analysis with secondary data from companies indexed in the Jakarta Islamic Index and Sharia Mutual Fund from year 2013 to 2017, it is found that both growth and value of stock have equally affected the profit of Sharia Mutual Funds. In addition, growth of stock has a larger R-Square than its value which means that the investors or fund managers would need to observe the stock growth more often than its value in order to predict future profitability of Shariah funds.  It is expected that the results of this study can provide additional insight to investment managers when choosing a portfolio for investors. For investors, this information is useful to predict the risk and return that they will receive from the investment.

scholarly journals Automatic Detection of Microsleep Episodes With Deep Learning

2021 ◽  
Vol 15 ◽  
Author(s):  
Alexander Malafeev ◽  
Anneke Hertig-Godeschalk ◽  
David R. Schreier ◽  
Jelena Skorucak ◽  
Johannes Mathis ◽  
...  
Keyword(s):  
Deep Learning ◽  
Short Term Memory ◽  
Internal Representation ◽  
Automatic Detection ◽  
Maintenance Of Wakefulness Test ◽  
Lstm Network ◽  
Definition Of ◽  
Artificial Neuronal Network ◽  
Electroencephalogram Eeg ◽  
Laborious Work

Brief fragments of sleep shorter than 15 s are defined as microsleep episodes (MSEs), often subjectively perceived as sleepiness. Their main characteristic is a slowing in frequency in the electroencephalogram (EEG), similar to stage N1 sleep according to standard criteria. The maintenance of wakefulness test (MWT) is often used in a clinical setting to assess vigilance. Scoring of the MWT in most sleep-wake centers is limited to classical definition of sleep (30 s epochs), and MSEs are mostly not considered in the absence of established scoring criteria defining MSEs but also because of the laborious work. We aimed for automatic detection of MSEs with machine learning, i.e., with deep learning based on raw EEG and EOG data as input. We analyzed MWT data of 76 patients. Experts visually scored wakefulness, and according to recently developed scoring criteria MSEs, microsleep episode candidates (MSEc), and episodes of drowsiness (ED). We implemented segmentation algorithms based on convolutional neural networks (CNNs) and a combination of a CNN with a long-short term memory (LSTM) network. A LSTM network is a type of a recurrent neural network which has a memory for past events and takes them into account. Data of 53 patients were used for training of the classifiers, 12 for validation and 11 for testing. Our algorithms showed a good performance close to human experts. The detection was very good for wakefulness and MSEs and poor for MSEc and ED, similar to the low inter-expert reliability for these borderline segments. We performed a visualization of the internal representation of the data by the artificial neuronal network performing best using t-distributed stochastic neighbor embedding (t-SNE). Visualization revealed that MSEs and wakefulness were mostly separable, though not entirely, and MSEc and ED largely intersected with the two main classes. We provide a proof of principle that it is feasible to reliably detect MSEs with deep neuronal networks based on raw EEG and EOG data with a performance close to that of human experts. The code of the algorithms (https://github.com/alexander-malafeev/microsleep-detection) and data (https://zenodo.org/record/3251716) are available.

scholarly journals Determinan Makro Ekonomi Yang Mempengaruhi Nilai Aktiva Bersih Reksadana Syariah

2021 ◽  
Vol 4 (2) ◽  
pp. 146
Author(s):  
Khoirunnisa Azzahra ◽  
Baiq Fitri Arianti
Keyword(s):  
Exchange Rate ◽  
Mutual Funds ◽  
Mutual Fund ◽  
Linear Regression Analysis ◽  
Sampling Technique ◽  
Total Sample ◽  
Macroeconomic Factors ◽  
Asset Value ◽  
Net Asset Value ◽  
Net Assets

The purpose of this study was to determine and analyze macroeconomic factors such as inflation, exchange rates and Indonesian Sharia bank certificates on Net Asset Value. Sources of data obtained from OJK and BI with 5 years of observation, The sampling technique used in this study is non-probability sampling, that is by using saturated sampling with a total sample of 60 data. The method used in this research is descriptive statistical analysis, classical assumption test, multiple linear regression analysis and hypothesis testing. By using the Statistical Package for the Social Science (SPSS) version 22.0 For Windows. The results of this study indicate that inflation has no effect on the value of net assets, while the exchange rate and SBIS partially affect the value of net assets. Simultaneously inflation, exchange rate and SBIS affect the net asset value. Net asset value (NAV) is important in mutual funds, because net asset value is one of the benchmarks in unifying mutual fund performance, the net asset value of equity/unit development mutual funds has increased, and vice versa decreased the value of initial mutual fund net assets/unit participation has decreased.

scholarly journals Classification of Covid-19 Coronavirus, Pneumonia and Healthy Lungs in CT Scans Using Q-Deformed Entropy and Deep Learning Features

Entropy ◽  
2020 ◽  
Vol 22 (5) ◽  
pp. 517 ◽  
Author(s):  
Ali M. Hasan ◽  
Mohammed M. AL-Jawad ◽  
Hamid A. Jalab ◽  
Hadil Shaiba ◽  
Rabha W. Ibrahim ◽  
...  
Keyword(s):  
Deep Learning ◽  
Short Term Memory ◽  
Accurate Method ◽  
Short Term ◽  
Lung Scan ◽  
Deep Learning Features ◽  
Long Short Term Memory ◽  
Lstm Network ◽  
Ct Slices

Many health systems over the world have collapsed due to limited capacity and a dramatic increase of suspected COVID-19 cases. What has emerged is the need for finding an efficient, quick and accurate method to mitigate the overloading of radiologists’ efforts to diagnose the suspected cases. This study presents the combination of deep learning of extracted features with the Q-deformed entropy handcrafted features for discriminating between COVID-19 coronavirus, pneumonia and healthy computed tomography (CT) lung scans. In this study, pre-processing is used to reduce the effect of intensity variations between CT slices. Then histogram thresholding is used to isolate the background of the CT lung scan. Each CT lung scan undergoes a feature extraction which involves deep learning and a Q-deformed entropy algorithm. The obtained features are classified using a long short-term memory (LSTM) neural network classifier. Subsequently, combining all extracted features significantly improves the performance of the LSTM network to precisely discriminate between COVID-19, pneumonia and healthy cases. The maximum achieved accuracy for classifying the collected dataset comprising 321 patients is 99.68%.

Sign in / Sign up

Export Citation Format

Share Document