Mutua Kilai

Financial time series are characterized by non-linearity, inherent volatility, and changing temporal patterns making accurate predictions very challenging. This study deploys a CNN-LSTM hybrid model to predict the USD/KES exchange rate giving particular emphasis to the effect the size of the historical input window has on the hybrid model forecasting performance. The study relied on Daily USD/KES exchange-rate data obtained from the Central Bank of Kenya to train, evaluate and validate the hybrid model. This data was prepossessed and transformed into sequential inputs using different window configurations, allowing the models to learn from varying amounts of historical information. The study combined a CNN-based local pattern extraction with an LSTM-based temporal dependency learning model to develop a hybrid CNN-LSTM model. The hybrid's model performance was then subjected to a comparison with standalone CNN and LSTM model and evaluated using Root Mean Square Error (RMSE) and Mean Absolute Error (MAE) The hybrid CNN-LSTM prediction results indicated that the model was more effective in prediction among the evaluated architecture. The choice of the input window sizes affected the model's ability to learn and capture trends in the time series data. The results from the rolling-window experiments demonstrated that shorter and moderately sized historical windows can provide a more responsive representation of recent exchange-rate movements than an expanding window, particularly during periods of pronounced market variation. These findings highlight the importance of considering both model architecture and input-window design when developing deep-learning approaches for USD/KES exchange-rate forecasting. The study provides an empirical basis for selecting appropriate historical window sizes when applying hybrid deep-learning models to volatile financial time series in the Kenyan foreign-exchange market.

This paper introduces the Marshall–Olkin Alpha Power Transformed Extended‐X (MOAPTE‐X) family of distributions, a highly flexible class designed to address limitations of existing models. We develop a novel four‐parameter distribution, called the Marshall–Olkin Alpha Power Transformed Extended‐Inverted Kumaraswamy (MOAPTE‐IK), with one scale parameter and three shape parameters, by extending the inverted Kumaraswamy distribution through the integration of the Marshall–Olkin method and alpha power transformations, thereby enhancing its capacity to model complex hazard rate functions. Some of the fundamental mathematical properties of the MOAPTE‐IK, including its quantile function, moments, generating functions, mode, order statistics, and Rényi entropy, are derived, establishing its theoretical robustness. Parameter estimation is performed using the maximum likelihood approach, and the behavior of the resulting estimates is assessed using Monte Carlo simulation, demonstrating consistency, accuracy, and efficiency. Comparative analyses using two real datasets reveal that the MOAPTE‐IK distribution consistently outperforms competing models such as Marshall–Olkin Extended Inverted Kumaraswamy (MOEIK) distribution, Alpha Power Inverted Kumaraswamy (APIK) distribution, and Inverted Kumaraswamy (IK) distribution. The suggested model consistently performs better than its competitors in terms of selection criteria (AIC, CAIC, BIC, HQIC) and goodness‐of‐fit tests (K‐S, A ^∗ , w ^∗ ). The proposed distribution offers a powerful, flexible tool for modeling complex data across multiple disciplines, including engineering, healthcare, and social sciences.

An intuitive and user-friendly package designed to aid undergraduate students in understanding and applying econometric methods in their studies, Tailored specifically for Econometrics and Regression Modeling courses, it provides a practical toolkit for modeling and analyzing econometric data with detailed inference capabilities.

In this study, an extension of XGamma distribution has been proposed and studied. The extension has an additional parameter accounting for the shape of the distribution. The properties of the proposed distribution were derived and discussed. The estimation of the parameters was done using the maximum likelihood method. The study's uniqueness is in developing a parametric regression model capable of competing with the classical regression model and also useful in the face of censored data. The applicability and flexibility were demonstrated using simulation studies and some lifetime data.

Using the power transformation method, we introduce a generalized version of the unit Gumbel type-2 distribution. The new lifetime distribution is called the power unit Gumbel type-2 distribution (PUGT2D). The new distribution’s statistical and reliability properties are given, and some estimation methods are proposed for estimating the model parameters. The usefulness and flexibility of the new distribution are illustrated with real datasets. Results based on log-likelihood, information statistics, and goodness-of-fit test results showed that the PUGT2D better fits the data than the other competing distributions. Moreover, a new regression model based on the new distribution is introduced and demonstrated to exhibit superior applicability through a numerical example.

In this paper, a bivariate power Lomax distribution based on Farlie-Gumbel-Morgenstern (FGM) copulas and univariate power Lomax distribution is proposed, which is referred to as BFGMPLx. It is a significant lifetime distribution for modeling bivariate lifetime data. The statistical properties of the proposed distribution, such as conditional distributions, conditional expectations, marginal distributions, moment-generating functions, product moments, positive quadrant dependence property, and Pearson’s correlation, have been studied. The reliability measures, such as the survival function, hazard rate function, mean residual life function, and vitality function, have also been discussed. The parameters of the model can be estimated through maximum likelihood and Bayesian estimation. Additionally, asymptotic confidence intervals and credible intervals of Bayesian’s highest posterior density are computed for the parameter model. Monte Carlo simulation analysis is used to estimate both the maximum likelihood and Bayesian estimators.

Streamflow prediction is vital to control the effects of floods and mitigation. Physical prediction model often provides satisfactory results, but these models require massive computational work and hydrogeomorphological variables to develop a prediction system. At the same time, data-driven prediction models are quick to apply, easy to handle, and reliable. This study investigates a new hybrid model, the wavelet bootstrap quadratic response surface, for accurate streamflow prediction. Wavelet analysis is a well-known time-frequency joint analysis technique applied in various fields like biological signals, vibration signals, and hydrological signals. The wavelet analysis is used to denoise the time series data. Bootstrap is a nonparametric method for removing uncertainty that uses an intensive resampling methodology with replacement. The authors analyzed the results of the studied models with different statistical metrics, and it has been observed that the wavelet bootstrap quadratic response surface model provides the most efficient results.

The objective of this study is to construct a new distribution known as the weighted Burr–Hatke distribution (WBHD). The PDF and CDF of the WBHD are derived in a closed form. Moments, incomplete moments, and the quantile function of the proposed distribution are derived mathematically. Eleven estimate techniques for estimating the distribution parameters are discussed, and numerical simulations are utilised to evaluate the various approaches using partial and overall rankings. According to the findings of this study, it is recommended that the maximum product of spacing (MPSE) estimator of the WBHD is the best estimator according to overall rank table. The actuarial measurements were derived to the suggested distribution. By contrasting the WBHD with other competitive distributions using two different actual data sets collected from the COVID-19 mortality rates, we show the importance and flexibility of the WBHD.

The exponentiated generalized Gull alpha power exponential distribution is an extension of the exponential distribution that can model data characterized by various shapes of the hazard function. However, change point problem has not been studied for this distribution. In this study, the change point detection of the parameters of the exponentiated generalized Gull alpha power exponential distribution is studied using the modified information criterion. In addition, the binary segmentation procedure is used to identify multiple change point locations. The assumption is that all the parameters of the EGGAPE distributions are considered changeable. Simulation study is conducted to illustrate the power of the modified information criterion in detecting change point in the parameters with different sample sizes. Three applications related to COVID-19 data are used to demonstrate the applicability of the MIC in detecting change point in real life scenario.

We steered comparative analysis of manifold supervised dimension reduction methods by assimilating customary multiobjective standard metrics and validated the comparative efficacy of supervised learning algorithms in reliance on data and sample complexity. The question of sample and data intricacy is deliberated in dependence on automating selection and user-purposed instances. Different dimension reduction techniques are responsive to different scales of measurement and supervision of learning is also discussed comprehensively. In line with the prospects, each technique validated diverse competence for different datasets and there was no mode to gauge the general ranking of methods trustily available. We especially engrossed the classifier ranking and concocted a system erected on weighted average rank called weighted mean rank risk adjusted model (WMRRAM) for consensus ranking of supervised learning classifier algorithms.