STATISTICAL MODEL FOR ESTIMATING THE POPULATION THAT RECEIVES HUMANITARIAN AID FOR DISASTER IN PERU

Authors

  • Santiago Salvador Montenegro-Canario Universidad Nacional Federico Villarreal. Escuela Universitaria de Post Grado. Lima, Perú. https://orcid.org/0000-0002-9306-3777
  • Napoleón Ambrocio-Barrios Universidad Nacional Federico Villarreal. Escuela Universitaria de Post Grado. Lima, Perú. https://orcid.org/0000-0001-8989-9232
  • José Iannacone 2Universidad Nacional Federico Villarreal, Facultad de Ciencias Naturales y Matemáticas, Escuela Profesional de Biología. Laboratorio de Ecología y Biodiversidad Animal. Lima, Perú. 3 Facultad de Ciencias Biológicas. Universidad Ricardo Palma. Lima, Perú. https://orcid.org/0000-0003-3699-4732

DOI:

https://doi.org/10.24039/rtb20151321556

Keywords:

Humanitarian assistance, injured population disaster, models of multiple linear regression, statistical model, Peru

Abstract

This research aims to design a statistical model to estimate the population receiving humanitarian aid to people affected by disasters in Peru, for which has been used ten variables: emergencies occurring in Peru (X1) injured population (X2) affected population (X3) homes destroyed (X4) homes affected (X5) hectares of crop losses (X6) hectares of crops affected (X7) the probabilities of occurrence of emergencies and disasters (X8) emergencies such phenomena occurred in Peru (X9) emergencies and natural regions (X10) The data are 52,327 records from 2003 to 2014 from the National Information System for Response and Rehabilitation (SINPAD), and the National Civil Defense Institute of Peru (INDECI) features on Web Platform, whose record of emergencies and
disasters is the responsibility of regional governments throughout Peru through the Emergency Operations Centers (COES). The data used in the design of the models were classified by quarters. To design the model the main hypotheses worked with eight variables, however only three variables best explained the model with the correlation coefficients (R), determination (R2), Durbin and Watson (D), F test validating the model and the t Student test, checking the validity, consistency and reliability of the parameters within the acceptable range. The multiple linear regression models to estimate the population receiving humanitarian disaster (y) was: y = 72455.731+ 0.417X2 + 0.405X3 - 3292452.345X8. Regression models were also designed by type  of phenomenon and selected variables were prioritized by Pareto rule, where 80% of the damage was caused by the 20% of phenomena, so by having a record of twenty phenomena, leaving for the design of three models the following phenomena: frost, floods and rain, and fire, taking significant models. For frost the regression model was y = 5025.805 + 0.614X3 + 0.811X2 - 198.3119X8. To design the model of Floods and rains was: y = - 0.145 + 0.109X2 + 0.966X3 - 0.114
X4 + 0.449X 8. Finally the design model for fires in Peru was: y = - 49.914 + 0.520X2 + 0.966 X3 +  24.573,03X8. There also was designed three models of multiple linear regressions for the natural regions coast, highlands and jungle. For the coast the model was: y = 28469.5 + 0.358 X2 + 0.768X3 -1.736.203.8 X8 the highlands model was: y = 13.803,888 + 0.891 X2 + 0.253 X3, and the  forest model: y = 909.070 + 0.420X2 + 0.430X3 . This statistical tool helps care for the disaster-  affected population in Peru.

Downloads

Download data is not yet available.

Published

2015-12-12

How to Cite

Montenegro-Canario, S. S., Ambrocio-Barrios, N., & Iannacone, J. (2015). STATISTICAL MODEL FOR ESTIMATING THE POPULATION THAT RECEIVES HUMANITARIAN AID FOR DISASTER IN PERU. The Biologist, 13(2), 375–390. https://doi.org/10.24039/rtb20151321556

Issue

Section

Original Articles