درخت حوزه‌های تخصصی

Predicting macroeconomic indicators is very important for policymakers and economists. Unemployment is one of the key indicators of macroeconomics that has adverse economic and social consequences. So far, many models have been proposed to predict this variable, but models in which accounting information was used to predict unemployment rate were ignored. The purpose of this paper is to investigate the relationship between aggregate cost stickiness, as one of the known variables in accounting, and unemployment rate. To this end, seasonal macro level time series data of Tehran Stock Exchange (TSE) and macroeconomic data are analyzed in two stages from 2008:2 to 2018:1. In the first stage, the relationship between these two variables is determined by specifying a linear regression model that is estimated using the OLS method. To investigate the predictive power of this model, the RMSE criterion was estimated in two scenarios with and without aggregate cost stickiness. Secondly, the reaction of the unemployment rate in response to a shock from aggregate cost stickiness is estimated by a Vector Autoregressive (VAR) model and the share of this variable is measured in the fluctuations of unemployment rate. The results show that aggregate cost stickiness improves the forecast of unemployment rate in the horizon previous. Also, the shock of aggregate cost stickiness explains about 6.5 percent of unemployment rate fluctuations.

Market manipulation remains the biggest concern of investors in today’s securities market. The development of technologies and complex trading algorithms seems to facilitate stock market manipulation and make it inevitable for regulators to use Deep Learning models to prevent manipulation. In this research, a Denoising GAN-based model has been designed. The proposed model (GAN-DAE4) consists of a three-layer encoder along with a 2-dimension encoder as the discriminator and a three-layer decoder as the generator. First, using statistical methods such as sequence, skewness, and kurtosis tests and some unsupervised learning methods such as Contextual Anomaly Detection (CAD) and some visual and graphical methods, the manipulated stocks have been detected in the Tehran Stock Exchange from 2015 to 2020; then GAN-DAE4 and some supervised deep learning models have been applied to the prepared data set. The results show that GAN-DAE4 outperformed other deep learning models (with F2-measure 73.71%) such as Decision Tree (C4.5), Random Forest, Neural Network, and Logistic Regression.

This research investigates the capital market's efficiency relative to the decreasing and increasing information the components of cash and accrual of the accounting earnings. In the accrual accounting system, accounting earnings includes two components of cash and accrual. Information about decreasing and increasing values of the normal and abnormal portion of the changes in financial assets as a cash component is compared with the information on the decreasing and increasing amounts of discretionary and non-discretionary accruals. The required data was extracted from the financial statements of the listed companies in Tehran Stock Exchange during the years 2003 to 2017. In order to estimate the research models from the regression with the combined data as well as the equations system, the simultaneous equations system with the seemingly unrelated regression (SUR) approach are utilized and then some experiments are implemented to evaluate the research hypotheses by using Mishkin’s test. The results reveal that the capital market is inefficient in terms of the increasing information (positive portion) of accrual and discretionary accruals and information on abnormal changes in financial assets (increasing and decreasing), but rather on the information (negative) of discretionary and non-discretionary accruals and abnormal changes in the financial assets (increasing and decreasing).

The purpose of the present study was to validate the Adaboost machine learning and probit regression in the prediction of Management's overconfidence at present and in the future. It also compares the predicted models obtained during the years 2012 to 2017. The samples of the research were the companies admitted to the Tehran Stock Exchange, (financial data of 1292 companies/year in total). Data collection in the theoretical part of the study benefitted from the content analysis international research paper in library method and for calculating the data's Excel software was used, and in order to test the research hypotheses, Matlab 2017 and Eviews10.0 were used. The empirical findings demonstrate that The Adaboost's algorithm nonlinear prediction model represents the highest power in learning and prediction (performance of this model) the managerial over-confidence for this year and the following year, proved to be better than the probit regression prediction model.

Multiplicity of the companies experiencing financial distress in different countries and as a consequence, their bankruptcy and the impacts on other companies have necessitated conducting research on methods of prediction of such conditions and also their effects on other companies in the market. In this regard, this research has investigated the financial distress spillover in the automobile supply chain companies. For doing so, the methods of default probability time series KMV and the distance from default of four supply chain companies of Iran Khodro and four supply chain companies of SAIPA were calculated. Then, the financial distress spillover in these two major companies was measured in separated models using multivariate GARCH model. The results of the default probability of Iran Khodro companies showed that the default probability with pause of Khodro on the default probability of supply chain companies was significant and negative in 10% level. The results for SAIPA supply chain companies revealed that the default probability with pause of Khaspa had an impact on default probability of Kaspa, Pask and Khazin in significance level of 10%.

Due to the rapid changes and developments caused by technology in the digital age and the growing importance of achieving sustainability for the survival of businesses, the need to create the capabilities needed to innovate in the business model of organizations has become inevitable. This research is trying to provide a model for business model innovation with the aim of sustainability by considering the external factor of digitalization (digital transformation) and the internal factor of dynamic capabilities. The method of this research is mixed so that in the first stage the qualitative method of Meta-Synthesis is used and in the second stage the quantitative method of Interpretive Structural Modeling (ISM) is used. In this research, based on the systematic review of previous researches in scientific databases, 402 related articles were identified and then 46 final articles were selected during the screening process. Based on the thematic analysis, a total of 4 main categories and 14 factors were extracted. In the second stage of the research, an interpretive structural model (ISM) is developed to show the inter‐relationship of different factors and their levels of importance in the food industry.   The 8-level hierarchy model of factors influencing the sustainable business model innovation in the digital age with a dynamic capabilities approach was proposed. Results show that a new component called network value has been added to the traditional business model components. Furthermore, the MICMAC approach has been used to categorise factors according to their driving power and dependence. finally, digital technologies and Sensing were found to be the factors with the highest independence and the highest driving power, and “Economic Sustainability” was identified as the factor with the highest dependence and the least driving power

Achieving e-commerce success is influenced by a variety of interrelated activities. Identifying, analyzing and prioritizing these activities enable decision-makers firstly to evaluate the current success state of Electronic Commerce (EC) businesses more accurately and then to develop an efficient improvement plan to allocate limited resources to achieve a higher level of success. In this research, a Fuzzy Cognitive Maps- analytical hierarchy process (FCM-AHP) approach is used to address this issue by (1) determining interrelationships between influential activities, (2) determining how much these activities influence each other and the overall success, and (3) prioritizing activities to develop a sound improvement plan. In this approach, the FCM technique is used to indicate all possible causal interrelationships between activities and consider all feedback loops between them. Then, the AHP technique is used to determine the contribution weights of activities to the overall success. The obtained management matrices that make it possible to categorize influential activities and the results of improvement scenarios help the decision-makers to develop the most efficient improvement plan and gain the maximum benefit from the allocation of limited resources. An empirical study is implemented to serve as an example to indicate how the approach works.

The most important aspects of image processing are image enhancement. The visual form of an image can be enhanced by using image enhancement techniques for better human interpretation. In this research paper we discuss an   outline and analysis of commonly used image enhancement techniques using finger vein image or personal authentication. Also, experiments are carried out to compare performance of various types of filters for removal of noise from the noisy images through evaluation performance parameter such as mean square error (MSE), peak signal to noise ratio (PSNR) values and structural similarity (SSIM). It was found that application of max filter technique ensures an improved quality of the finger vein image. The mean filter is most advanced in de-noising the images. Mean filter is most efficient in eliminating the salt and pepper noise. From the experiments performed on finger vein image using SDUMLA-HMT database, it is proven that Weiner filters are outstanding for elimination of Gaussian Speckle and Poisson noises and thus, Weiner filter is found to be most appropriate and well-suited for eliminating nearly all types of noise.