Department of Statistics & Actuarial Sciencehttp://repository.dkut.ac.ke:8080/xmlui/handle/123456789/2032024-03-28T19:51:21Z2024-03-28T19:51:21ZA hybrid neural network model based on transfer learning for Arabic sentiment analysis of customer satisfactionBakhit, Duha Mohamed AdamNderu, LawrenceNgunyi, Antonyhttp://repository.dkut.ac.ke:8080/xmlui/handle/123456789/84862024-03-22T08:22:34Z2024-02-01T00:00:00ZA hybrid neural network model based on transfer learning for Arabic sentiment analysis of customer satisfaction
Bakhit, Duha Mohamed Adam; Nderu, Lawrence; Ngunyi, Antony
Sentiment analysis, a method used to classify textual content into positive, negative,
or neutral sentiments, is commonly applied to data from social media
platforms. Arabic, an official language of the United Nations, presents unique
challenges for sentiment analysis due to its complex morphology and dialectal
diversity. Compared to English, research on Arabic sentiment analysis is
relatively scarce. Transfer learning, which applies the knowledge learned from
one domain to another, can address the limitations of training time and computational
resources. However, the development of transfer learning for Arabic
sentiment analysis is still underdeveloped. In this study, we develop a
new hybrid model, RNN-BiLSTM, which merges recurrent neural networks
(RNN) and bidirectional long short-term memory (BiLSTM) networks. We used
Arabic bidirectional encoder representations from transformers (AraBERT), a
state-of-the-art Arabic language pre-trained transformer-based model, to generate
word-embedding vectors. The RNN-BiLSTM model integrates the strengths
of RNN and BiLSTM, including the ability to learn sequential dependencies
and bidirectional context. We trained the RNN-BiLSTM model on the source
domain, specifically the Arabic reviews dataset (ARD). The RNN-BiLSTMmodel
outperforms the RNN and BiLSTM models with default parameters, achieving
an accuracy of 95.75%. We further applied transfer learning to the RNN-BiLSTM
model by fine-tuning its parameters using random search. We compared the
performance of the fine-tuned RNN-BiLSTMmodel with the RNN and BiLSTM
models on two target domain datasets: ASTD and Aracust. The results showed
that the fine-tuned RNN-BiLSTM model is more effective for transfer learning,
achieving an accuracy of 95.44% and 96.19% on the ASTD and Aracust datasets,
respectively.
2024-02-01T00:00:00ZUnraveling Market Inefficiencies: Weak Arbitrage and the Information-Based Model for Option PricingOdin, MatabelAduda, Jane AkinyiOmari, Cyprian Ondiekihttp://repository.dkut.ac.ke:8080/xmlui/handle/123456789/83152023-11-30T09:48:09Z2023-11-01T00:00:00ZUnraveling Market Inefficiencies: Weak Arbitrage and the Information-Based Model for Option Pricing
Odin, Matabel; Aduda, Jane Akinyi; Omari, Cyprian Ondieki
Discrepancies between theoretical option pricing models and actual market
prices create arbitrage opportunities in financial markets. Despite being
widely used in option pricing, the famous Black-Scholes model estimates op-
tion values based on the strict assumption of no arbitrage. In addition, its as-
sumptions of constant volatility and log-normal asset price distribution may
not fully capture real-world market dynamics, resulting in mispricing and
potential arbitrage opportunities. The Information-based model is adopted as
an alternative to address this, allowing for stochastic volatility, non-specific
asset price distributions, and variable transaction costs. This study extends
the IBM by developing a pricing equation incorporating weak arbitrage pos-
sibilities using the weaker form of no-arbitrage termed as the Zero Curvature
condition. The equation incorporates an adjusted risk-free rate, influenced by
an arbitrage measure and option derivatives. Empirical findings based on the
iShares S&P 100 ETF American call options dataset demonstrate that captur-
ing weak arbitrage improves theoretical option price estimates, reducing dis-
crepancies and potential arbitrage opportunities. Further research can focus
on validating and enhancing the Information-based model using alternative
financial assets data.
2023-11-01T00:00:00ZHierarchical Logistic Regression Model for Multilevel Analysis on the Uptake of Health Insurance in Nouakchott, MauritaniaTourad, Tourad CheikhNgunyi, AntonyImboga, Herberthttp://repository.dkut.ac.ke:8080/xmlui/handle/123456789/82472023-10-02T10:43:28Z2022-05-01T00:00:00ZHierarchical Logistic Regression Model for Multilevel Analysis on the Uptake of Health Insurance in Nouakchott, Mauritania
Tourad, Tourad Cheikh; Ngunyi, Antony; Imboga, Herbert
The availability of these complex statistical methods challenges public health researchers to articulate theories of
the causes of health behaviour that bring together factors defined at different levels. This study seeks to discuss the
hierarchical logistic regression model for multilevel analysis and test its application in analysing the uptake of health
insurance in Mauratania. The specific objectives of this study are to develop the hierarchical logistic regression model,
estimate the model parameters of the hierarchical logistic regression model, derive the maximum likelihood estimators of the
parameters of the hierarchical logistic regression model and apply the estimation procedure for the uptake of health insurance
data from Nouakchott, Mauritania. The study adopted an explanatory study design using secondary data obtained from
National Health Insurance funds in Mauritania. The hierarchical logistic regression model for multilevel analysis was used in
analysing the data. The analysed data is presented using the table. The obtained model can be used to predict the uptake
2022-05-01T00:00:00ZA Jump Diffusion Model with Fast Mean-Reverting Stochastic Volatility for Pricing Vulnerable OptionsNthiwa, Joy K.Kube, Ananda O.Omari, Cyprian O.http://repository.dkut.ac.ke:8080/xmlui/handle/123456789/82072023-09-29T09:44:07Z2023-09-01T00:00:00ZA Jump Diffusion Model with Fast Mean-Reverting Stochastic Volatility for Pricing Vulnerable Options
Nthiwa, Joy K.; Kube, Ananda O.; Omari, Cyprian O.
Te Black–Scholes–Merton option pricing model is a classical approach that assumes that the underlying asset prices follow
a normal distribution with constant volatility. However, this assumption is often violated in real-world fnancial markets, resulting
in mispricing and inaccurate hedging strategies for options. Such discrepancies may result into fnancial losses for investors and
other related market inefciencies. To address this issue, this study proposes a jump difusion model with fast mean-reverting
stochastic volatility to capture the impact of market price jumps on vulnerable options. Te performance of the proposed model
was compared under three diferent error distributions: normal, Student-t, and skewed Student-t, and under diferent market
scenarios that consist of bullish, bearish, and neutral markets. In a simulation study, the results show that our model under skewed
Student-t distribution performs better in pricing vulnerable options than the rest under diferent market scenarios. Our proposed
model was ftted to S&P 500 Index by maximum likelihood estimation for the mean and volatility processes and Gillespie
algorithm for the jump process. Te best model was selected based on AIC and BIC. Samples of the simulated values were
compared with the S&P 500 values and MSE computed at various sample sizes. Values of MSE at diferent sample sizes indicate
signifcant decrease to actual MSE values demonstrating that it provides the best ft for modeling vulnerable options.
2023-09-01T00:00:00Z