Precision and Performance: Regression Methods in Data Processing | Chapter 3 | Engineering Research: Perspectives on Recent Advances Vol. 8

Regression analysis is a fundamental statistical technique used for modelling relationships between multiple variables, playing a significant role in predictive analytics and artificial intelligence. It helps in evaluating dependencies between a dependent variable and one or more independent variables, making it essential for forecasting outcomes. This study focuses on the application of regression models to analyse vehicle dynamics. By utilising data such as traffic velocity, road gradient, and actual speed, we aim to predict a vehicle’s velocity profile. Various regression techniques—including linear regression, multivariate linear regression, and nonlinear regression—are examined to determine their effectiveness in data processing. The core objective of this research is to develop reusable functions for each model, eliminating dependence on predefined programming functions, while enabling effective visualisation of data for optimal model selection. One advantage of a regression model over factor or cluster analysis is that the regression model can be used to obtain an estimate of the actual amount of change in a dependent variable that occurs as a result of a change in an independent variable. This study only focuses on simple linear regression; future work should focus on the use of multiple regression to capture more complex relationships.

 

Author (s) Details

Sandhya Tatekalva
Department of Computer Science, S.V. University, Tirupati, India.

 

Please see the book here:- https://doi.org/10.9734/bpi/erpra/v8/5675

 

Perceived Parental Behaviour and Religiosity as Predictors of In-school Adolescent Sexual Risk Behaviour in Ibadan Metropolis, Nigeria | Chapter 8 | Medical Science: Trends and Innovations Vol. 1

Sexual risk behavior is an important contemporary concern, especially for the self-destruction and societal ills it causes. Research has been ongoing on adolescent sexual risk behavior in the country and recent studies still report high rates of pre-marital sexual activities among Nigerian adolescents. Sexual risk behavior among youth, defined as early sexual initiation, unprotected intercourse, or sex with multiple partners, is a major area of concern to parents or caregivers and health professionals given the many associated negative consequences. There is a dearth of literature on the interaction influence of perceived parental behavior and religiosity on adolescent sexual risk behavior in the country of Nigeria. This study, therefore, investigated the prediction of sexually risky behavior of the youth by their perception of their parental behaviors, and their religious belief. The roles of a few sociodemographic factors were also examined. A cross-sectional research design was employed. Using a multistage sampling technique, 1,589 participants, male (n=753), and female (n=836) aged 13 to 19 years were drawn from 10 private and 19 public secondary schools in three of the five Local Government Areas (LGA) in Ibadan Metropolis as participants for the study. Regression analyses revealed significant prediction by parental psychological control (β=.10, t=3.99, p<.01) while the other parental behavioral dimensions were not significant. Religiosity also significantly predicted sexual risk behavior among adolescents (F (1, 1584) =50.25, p<.001) (β=-0.18, t=-7.08, p<.001) indicating an increase in sexually risky behavior with a decline in religiosity. Age, Gender and family structure were also independent and significant predictors (β = 0.12, t=-4.89, p< 01; β= 0.30, t=-12.62, p<.01, β = 0.05, t=2.12, p<,05) respectively, implying the younger, male adolescents from polygamous setting are higher on sexual risk behavior. Implication is an overuse of psychological control. Efforts should be geared towards eliciting healthy parental behavior through teaching parenting skills and also giving adolescents cognitive therapy. Religiosity should also be encouraged.

 

Author (s) Details

Aderonke A. Akintola
Redeemers’ University, Ede, Osun State, Nigeria.

 

Please see the book here:- https://doi.org/10.9734/bpi/msti/v1/3749

Statistical Downscaling of Temperature and Precipitation for Mount Makulu, Zambia Using the Statistical Downscaling Portal for 2010-2099 | Chapter 3 | Emerging Issues in Environment, Geography and Earth Science Vol. 8

 Statistical downscaling of temperature and precipitation bridges the gap between the predictors and predictand and assist policymakers, researcher, and end users to assess the likely impacts of climate change on key sectors. This study simulated the future change in temperature and precipitation for Mount Makulu, Zambia using ERA-Interim reanalysis data within the Statistical Downscaling Portal (SDP). Climate scenarios were generated for temperature and precipitation for 2010-2039/1971-2000, 2040-2069/1971-2000 and 2070-2099/1971-2000 for RCP4.5 and RCP8.5. The Mann-Kendall test showed a significant positive trend for maximum and average temperature for the current and future climate scenarios at p<0.05. Results showed a higher variability and decreasing trend in annual precipitation under RCP 8.5 compared to RCP 4.5. The mean temperature change relative to the baseline would be 1.03oC, 1.21oC, 1.65oC, 1.87oC, 1.89oC and 2.23oC under RCP4.5 (2020), RCP8.5 (2020), RCP4.5 (2050), RCP8.5 (2050), RCP4.5 (2080) and RCP8.5 (2080), respectively. The use of PP improves the coarse spatial resolution of temperature and precipitation in GCMs and assists decision makers and end users in understanding the likely impacts of climate variability and change. The future climate scenarios under RCPs (RCP 4.5 and RCP 8.5) provides new insights in rainfall and temperature trends from 2010-2099. The findings provide useful information as inputs into the Sectors and National Adaptation Plans.

Author(s) Details:

Charles Bwalya Chisanga,
The Copperbelt University, SNR, PES, P. O. Box 21692, Kitwe, Copperbelt, Zambia.

Elijah Phiri,
University of Zambia, School of Agricultural Sciences, Department of Soil Science, P. O. Box 32379, Lusaka, Zambia.

Vernon R. N. Chinene,
University of Zambia, School of Agricultural Sciences, Department of Soil Science, P. O. Box 32379, Lusaka, Zambia.

Monday Chota,
Ministry of Education, P. O. Box 440045, Isoka, Muchinga, Zambia.

Edson Nkonde,
Zambia Meteorological Department, P. O. Box 30200, Lusaka, Zambia.

Please see the link here: https://stm.bookpi.org/EIEGES-V8/article/view/13615

Keywords: Scenarios, climate change, GCMs, perfect prognosis, statistical downscaling, SDP, predictand, predictor