Predicting energy poverty with combinations of remote-sensing and socioeconomic survey data in India: Evidence from machine learning

PurposeThe study uses machine learning techniques to cluster regional retail egg prices after 2000 in China. Furthermore, it combines machine learning results with econometric models to study determinants of cluster affiliation. Eggs are an inexpensiv, nutritious and sustainable animal food. Contextually, China is the largest country in the world in terms of both egg production and consumption. Regional clustering can help governments to imporve the precision of price policies and help producers make better investment decisions. The results are purely driven by data.Design/methodology/approachThe study introduces dynamic time warping (DTW) algorithm which takes into account time series properties to analyze provincial egg prices in China. The results are compared with several other algorithms, such as TADPole. DTW is superior, though it is computationally expensive. After the clustering, a multinomial logit model is run to study the determinants of cluster affiliation.FindingsThe study identified three clusters. The first cluster including 12 provinces and the second cluster including 2 provinces are the main egg production provinces and their neighboring provinces in China. The third cluster is mainly egg importing regions. Clusters 1 and 2 have higher price volatility. The authors confirm that due to transaction costs, the importing areas may have less price volatility.Practical implicationsThe machine learning techniques could help governments make more precise policies and help producers make better investment decisions.Originality/valueThis is the first paper to use machine learning techniques to cluster food prices. It also combines machine learning and econometric models to better study price dynamics.

show abstract

“…, 2022) and facilitated access to unavailable data for analysis (Graskemper et al. , 2021, 2022; Wang et al. , 2021a).…”

Section: Literature Reviewmentioning

confidence: 99%

Food price dynamics and regional clusters: machine learning analysis of egg prices in China

Chang

Zhou

Höschle

et al. 2022

CAER

Self Cite

View full text Add to dashboard Cite

show abstract

“…Methodologically, the study employs the Least Absolute Shrinkage and Selection Operator (LASSO) algorithm, which is a machine learning technique for features selection (Maruejols et al, 2022) which has recently gained increasing attention in the field of farmers´ behavior studies (Graskemper et al, 2021(Graskemper et al, , 2022Wang et al 2021). Robustness checks are realized through the application of cross-validated shrinkage penalties.…”

Section: Introductionmentioning

confidence: 99%

Participation in a mutual fund covering losses due to pest infestation: analyzing key predictors of farmers’ interest through machine learning

Höschle

Trestini

Giampietri

2023

IFAM

View full text Add to dashboard Cite

In the context of intensified Halyomorpha halys infestations in Italy, this paper provides a very first investigation of key factors that drive fruit growers’ intention to participate in a mutual fund (MF) compensating production losses due to this invasive insect. Data were collected in Veneto Region in Italy, where many farmers suffered H. halys attacks, and interest in the development of innovative risk management tools is growing. The study investigates how behavioral (risk attitude, risk perception) and personality factors (self-efficacy, locus of control) explain farmers’ intention to participate in the MF, additionally controlling for a large number of primary control data (e.g. farmers’ perceptions and characteristics, farm characteristics). The study assumes approximate sparsity and applies the least absolute shrinkage and selection operator (LASSO), a machine learning technique which represents an original approach for research on risk management. Our empirical analysis reveals that farmers’ intention to participate in the MF is driven by an interplay between the perceived risk of production loss, the benefits from participation in the fund, and the farm age, rather than by socio-economic characteristics of the farm. Results provide valuable insights for policymakers and local stakeholders to implement a mutual fund close to the farmers’ needs.

show abstract

“…Non-linearity of the price has been found to affect results of elasticities, but to be too small to matter (Henson, 1984), albeit for developed countries. In the context of a low-income country, energy poverty and affordability are major policy concerns (Jain et al, 2018;Wang, Maruejols, and Yu, 2021), and families spend larger shares of their budget on energy expenditures. This study is among the firsts which attempt to quantify the income effect bias resulting from smaller rates (Sun and Lin, 2013;Khanna et al, 2016).…”

Section: Discussionmentioning

confidence: 99%

“…It also has consequences on deforestation of certain areas and global warming (Masera et al, 2015). In this domain, literature on energy poverty identifies that, given the recent progress in rural electrification (almost 100 percent households are now connected to an electricity source in India, China, Vietnam…), the main obstacles lie in the hours of electricity supply available per day and its affordability for poor households (Agarwal et al, 2020;H. Wang et al, 2021).…”

Section: Research Questionmentioning

confidence: 99%

“…Ownership of small appliances however is not found to have sufficient association with electricity consumption and expenditure. Among socio-economic variables, it is not surprising to find that household size is selected by the LASSO approach as it has often been associated with energy consumption (Ozughalu and Ogwumike, 2019;Sharma et al, 2019;H. Wang et al, 2021).…”

Section: Model Selectionmentioning

confidence: 99%

See 1 more Smart Citation

Economic Implications of Changing Household Management of Basic Needs for Poverty Alleviation and Sustainability Policy, an Illustration of Food and Energy

Maruejols¹

View full text Add to dashboard Cite

The sharp reduction of extreme poverty worldwide and improvement to living standards in the last 20 years are remarkable and imply that a large number of people in developing countries are able to afford more than their minimum subsistence. However, without adequate protection mechanisms, this progress remains fragile and can easily be reverted, as seen by the devastating socio-economic effects of the COVID-19 pandemic restrictions and disturbances to the economies. Households situated just above the poverty line remain vulnerable, which points to the evident need for continuous public support to secure their position out of poverty. However, as income increases sufficiently to exit extreme poverty, households' preferences shift and their strategies to fulfill their basic needs as well. These modifications have implications for the design of appropriate support programs, to help individuals stabilize and improve their quality of life, but also encourage sustainable choices.This dissertation expands knowledge on mechanisms of household consumption and fulfillment of basic necessities, at various levels of income. Specifically, food and energy consumption are examined, as they are evidently required for basic human development and well-being, but their consumption patterns change noticeably with income, to better quality diets and to modern energy sources respectively. Accounting for these changes is a condition for the success of interventions to secure decent living standards and sustainability. Specifically, the four essays of this work provide illustrations for rural populations of selected Asian countries and considers the choice of income versus non-income forms of support under different settings.In particular, essay 1 examines the changing preferences of households for home-produced food over market purchases at various income levels. A theoretical framework places this arbitrage under the constraints of minimum food and minimum cash requirements to be fulfilled for survival. A nonlinear relationship appears between the share of home food in total food expenditure and income. Therefore, contrary to the view that home food is associated with subsistence, the share of home-produced food does not always shrink with better incomes.Using data from rural India, a semiparametric analysis shows the pattern is stronger for households whose main activity is not farming. Analysis of market changes or policy intervention on food consumption patterns should acknowledge that higher home-food consumption may not always represent a sign of impoverishment.

show abstract

Predicting energy poverty with combinations of remote-sensing and socioeconomic survey data in India: Evidence from machine learning

Cited by 24 publications

References 67 publications

Food price dynamics and regional clusters: machine learning analysis of egg prices in China

Food price dynamics and regional clusters: machine learning analysis of egg prices in China

Participation in a mutual fund covering losses due to pest infestation: analyzing key predictors of farmers’ interest through machine learning

Economic Implications of Changing Household Management of Basic Needs for Poverty Alleviation and Sustainability Policy, an Illustration of Food and Energy

Contact Info

Product

Resources

About