Purpose
– Making decisions in finance have been regarded as one of the biggest challenges in the modern economy today; especially, analysing and forecasting unstable data patterns with limited sample observations under the numerous economic policies and reforms. The purpose of this paper is to propose suitable forecasting approach based on grey methods in short-term predictions.
Design/methodology/approach
– High volatile fluctuations with instability patterns are the common phenomenon in the Colombo Stock Exchange (CSE), Sri Lanka. As a subset of the literature, very few studies have been focused to find the short-term forecastings in CSE. So, the current study mainly attempted to understand the trends and suitable forecasting model in order to predict the future behaviours in CSE during the period from October 2014 to March 2015. As a result of non-stationary behavioural patterns over the period of time, the grey operational models namely GM(1,1), GM(2,1), grey Verhulst and non-linear grey Bernoulli model were used as a comparison purpose.
Findings
– The results disclosed that, grey prediction models generate smaller forecasting errors than traditional time series approach for limited data forecastings.
Practical implications
– Finally, the authors strongly believed that, it could be better to use the improved grey hybrid methodology algorithms in real world model approaches.
Originality/value
– However, for the large sample of data forecasting under the normality assumptions, the traditional time series methodologies are more suitable than grey methodologies; especially GM(1,1) give some dramatically unsuccessful results than auto regressive intergrated moving average in model pre-post stage.
Social media have become very popular in the last few decades. Users rely on social network sites like Twitter, Facebook, YouTube, and LinkedIn for both information and entertainment needs. Social media analytics with data mining technology could be an analysis axis centered on extracting trends, patterns, and rules from the social media pool, to serve the people and organizations to have optimum choices concerning many disciplines. The traditional media analytical techniques appear obsolete and inadequate to gratify this immense array of unstructured social media knowledge characterized by three key problems namely; size, noise, and dynamism, predominantly shifting from the batch scale to the streaming one. The objective of this study is to investigate the data mining techniques that were used by social media networks during the years 2010 and 2020. The effort is a systematic review of content analysis in studies within the field of social media analytics that was published in principal databases. 125 articles were reviewed in this paper. Content analysis was implemented based on their approach, tools utilized, language, the dataset used, country, year, and nature of the experiment. The review discovered that 22 data mining techniques were employed with social media data while frequently used in Artificial Neural Network (ANN), Bayesian networks (BN) and Support Vector Machine (SVM), K-means Clustering, and Neuro-Fuzzy Logic Approach. The study has focused to assist the involved analyzers and educators to capture the research trends and problems associated with the Social media analytics process with future research initiatives.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.