Multivariate statistics of biogas and national grid energy consumers for countries with limited and scarce data: Nepal as a case study

Abstract

This paper analyzes the energy consumption pattern of rural households with respect to multiple variables. Some of these variables are distance travelled for the collection of firewood, firewood saved, sources of firewood collection, type of house, location and type of latrine used, no. of residents more than 15 years of age and composition of adult women more than 25% of the total family size. A survey of 400 households of biogas users and 300 households of national grid energy users of Nepal generated more than 350 multivariate data. Multivariate statistics is used here to quantify and explain the relationship between these variables. Firstly from the factor analysis, three factors are identified. These three factors can explain 78% of the variance of the energy consumption pattern of rural households. Then, multivariate analysis of variance is also conducted. After this, multinomial logistic regression is used for predicting benefits of biogas (time saved) on data 400 households of biogas users. In this, the probability of change in time saved is predicted for a unit change in distance travelled in the collection of firewood before and after the construction of plant, reduced pollution, reduced fuel expenses and amount of firewood saved. Thus, evidence-based analysis of dynamics of improvement of people in general and women in particular when biogas is used as an energy source for cooking is done. The results obtained here can be used for other least developed country in Asia and Africa having limited and scarce official records.