Spatial Probit Model Research Articles

Analysis of Female Employment Rates in Türkiye with Duration Variable Spatial Logit and Duration Variable Spatial Probit Models

Female employment holds significant importance in achieving sustainable economic growth. In this regard, initiatives aimed at increasing female employment are among the fundamental goals of countries. Studies on female employment are expected to assist in developing strategies and policies that guide countries in reaching their relevant objectives. This study aims to investigate the spatial interactions and determinants of female employment rates in Level-2 regions in Türkiye during the 2018-2020 period using duration variable spatial logit and duration variable spatial probit models. The analysis revealed findings indicating a positive impact on education levels and a negative impact of the Gini coefficient and the duration variable on female employment rates. The findings related to the duration variable suggest a decreasing likelihood of female employment rates being above the median value during the 2018-2020 period.

Peer effects affect the adaptation behaviour of rural residents to geohazards in China: a new policy pathway

ABSTRACT Disaster prevention and mitigation (DPM) is one of the responses to extreme climate change impacts and depends on the adoption of risk adaptation behaviour by rural residents. While residents exhibit a high willingness to adapt, they also show low behaviour with reference to DPM. While adaptation behaviour of individuals has been studied extensively, the mutual impact of adaptation behaviour among residents has so far been neglected in the literature. Evaluating this interaction could serve as an important foundation for refining disaster reduction policies. Drawing on social cognitive theory, we propose a novel theoretical framework to analyze peer effects on individual decision-making in DPM. Using data gathered from 536 respondents in Sichuan Province, China, this paper adopts the Bayes Markov chain Monte Carlo estimation method to perform a regression of the spatial probit model. Further, the propensity score matching model was used to explore peer effects on the transformation between residents’ willingness to actual behaviour. Results indicate significant and robust peer effects on DPM behaviour among residents. A heterogeneity test indicated that peer effects are predominantly observed among residents with low and medium social status perception and not in those with high social status perception. Peer effects have a significant and positive influence on the transformation from residents’ willingness to their behaviour. Residents are not completely independent in their decision-making, and complex social mechanisms dictate rational and irrational choices of behaviour. Therefore, it is necessary to fully consider positive peer effects on residents in the formulation of policies. Key policy insights Focusing on the adaptation behaviours and peer effects of rural residents is conducive to improving the public's acceptance of disaster reduction policies. It is difficult for homogenization policies to accurately meet the needs of rural residents in China, and the response to geohazard events requires diversified local policies. Through peer effect, the government can create a kind of environmental soft pressure and promote the transformation of residents’ willingness into behaviour. Due to peer effects, the government can utilize the leading role of a small number of target groups to reduce the cost of implementing disaster reduction policies.

Identifying the determinants of E-commerce offerings among grocery store locations in the United States: A spatial econometric analysis

This paper leverages store location data and analysis software from retail industry vendors to analyze the unique geographies of electronic commerce (e-commerce) offerings among major grocery chains in the United States. Join-Count statistics are calculated to identify patterns of clustering among offerings, and a spatial probit model is implemented to regress various trade area characteristics on a given grocery banner's decision to offer some form of e-commerce (pickup, delivery, or both) or no offering at a given store location. A regional case study of the Southeastern United States demonstrates how model results can shift based on scale, pointing to unique regional dynamics that deviate from broader national trends. This analysis carries implications for location planning and competitive defense strategies in the grocery industry, and also establishes avenues for future research.

Identifying effective measures to enhance the recovery of small and micro enterprises after multiple waves of COVID-19 – A case study from Dongguan, China

The ability of a small and micro enterprise (SME) to recovery after a pandemic outbreak can directly affect national economic growth and sustainable development (Sustainable Development Goal 11). Thus, the current study quantitatively identifies, based on the spatial probit model, effective measures that enhance the performance and survival of SMEs after multiple waves of shocks induced by COVID-19. Additionally, this work explores the spatial correlation for post-pandemic recovery performance of businesses adopting a novel approach. Two field investigations were conducted in Dongguan City, in China, where questionnaires were released to 592 SMEs over a two-year period. The results obtained showed that the overall recovery performance of SMEs after the pandemic showed a positive spatial correlation, while this correlation varied at different local regions. More in detail, self-media marketing, borrowing money from family or friends, requesting bank loans and tax relief have been identified to be the key measures to effectively support the recovery and the increase chances for small businesses to remain operative after the pandemic. However, the success of these measures varies across businesses with different characteristics (owner's age and industry experience, primary market and business size, pre-pandemic financial condition). It is then suggested based on the results obtained that policy managers should formulate differentiated policies in terms of assistance measures for businesses targeting dissimilar characteristics as well as the needs of different regions, because those may have been impacted differently by the pandemic.

Neighborhood Does Matter: Farmers’ Local Social Interactions and Land Rental Behaviors in China

The transfer of farmland is an important area of rural development research; however, the impact of rural social networks has been neglected in studies. The aim of this study is to explore the effects, mechanisms, and heterogeneity of neighbors’ behavior on the process of land renting by farmers. Based on the data of the China Family Panel Studies in 2018, this research empirically analyzes the impact of community-level, local social interactions on the land rental behavior of farmers and its mechanisms using a spatial probit model. The results of this study indicate that neighbors’ land rental behavior positively and significantly affects that of other farmers in the same village. In addition, neighbors’ land rental encourages other farmers in the same village to follow suit through an increase in the perceived importance of the Internet among the farmers. In addition, there is heterogeneity in neighborhood influence. Notably, the impact of social networks on the renting out of the land by farmers, as evidenced in this study, is a key factor in accelerating the circulation of rural land and promoting rural development, thus contributing to the process of rural revitalization and its recording in the literature.

Modeling drivers of farming system trajectories in Mediterranean peri-urban regions: Two case studies in Avignon (France) and Pisa (Italy)

CONTEXTMediterranean peri-urban regions have changed considerably in recent decades, leading to different farming system trajectories. The high level of heterogeneity and complexity of these areas are particularly challenging in terms of understanding the ongoing processes and their drivers, including the effects of spatial interactions, in order to make an initial assessment of the possible levers for territorial management. OBJECTIVEWe aim to show a new modeling approach to assess the factors explaining farming system trajectories on Mediterranean peri-urban areas, including an analysis of the spillover from neighboring areas, which is often neglected in these kinds of methods. The model was tested in two urban regions, located in Pisa (Italy) and Avignon (France), leading to very different perspectives, given the differences in terms of farming systems and environmental characteristics. Our model also highlights the factors acting on peri-urban farming system trajectories. METHODSWe first analyzed the farming systems characteristics and their trajectories, based on a land parcel identification system database and agricultural census. Then, we estimated explanatory factors at the farm level applying a spatially explicit model which takes into spatial autocorrelations into account: a Spatial Autoregressive Probit model. RESULTS AND CONCLUSIONSOur results provide new evidence supporting the hypothesis that peri-urban farming systems trajectories are influenced by other variables than the distance to the city center and the urbanization processes. Although urban sprawl characterizes medium-sized regions, this factor is not relevant on farming systems trajectories in the region of Pisa, where structural agro-pedoclimatic drivers contribute more significantly to the probability of change and, therefore, highlighting the greater stability of these farming systems. The region of Avignon reveals different aspects: a dense road infrastructure and favorable environmental conditions (availability of irrigation water during warm periods, adequate temperatures in winter, deep soils) increase the probability of intensification within the farming system. SIGNIFICANCEOur methodological approach, which is based on a spatial probit model, measures the effect of the spatial dependence of the dependent variables. In addition, the preliminary classification in terms of farming systems improves our understanding of the land trajectories. The results obtained provide new insights into the processes underway in Mediterranean peri-urban areas.

Fitting spatial autoregressive logit and probit models using Stata: The spatbinary command

Starting from version 15, Stata allows users to manage data and fit regressions accounting for spatial relationships through the sp commands. Spatial regressions can be estimated using the spregress, spxtregress, and spivregress commands. These commands allow users to fit spatial autoregressive models in cross-sectional and panel data. However, they are designed to estimate regressions with continuous dependent variables. Although binary spatial regressions are important in applied econometrics, they cannot be estimated in Stata. Therefore, I introduce spatbinary, a Stata command that allows users to fit spatial logit and probit models.

Modeling Local Variations in Intermarriage

We utilized all Spanish marriage records available at the municipality level from 2005–2007 to model spatial variations in intermarriage. We constructed a spatial regime zero-inflated Poisson model and grouped-data probit model, with spatially lagged regressors, to predict the absolute and relative presence of intermarriage between Spaniards and migrants based on structural characteristics of the local marriage markets and their neighboring areas (i.e., relative group size, homogeneity of national origins, and sex ratio indicators). Our models do not assume collapsibility of the marriage market. Instead, they incorporate the local dimension of the marriage market and examine the association between intermarriage and structural variables at the spatial local level. The model also investigates intermarriage variation by size of place. The local characteristics of the marriage markets are robust indicators of both the absolute and relative importance of intermarriage, but their impact varies by size of municipality. The relative size of the migrant community positively impacts intermarriage. The homogeneity of the origins of migrants is negatively related to it. The impact of sex ratios in the migrant and native communities on intermarriage is not uniform across all municipalities and is not always related to more intermarriage.

Modelling avocado-driven deforestation in Michoacán, Mexico

As demand for avocado climbs, avocado production in Michoacán—Mexico’s biggest avocado growing region—expands into new places. We use a spatial probit model to project the geographic distribution of likely future avocado expansion and analyze those results to determine (a) threats to specific forest types and (b) how the distribution of avocado is shifting spatially under current and future climate scenarios. Our results suggest that avocado expansion in Michoacán is strongly driven by distance to existing agriculture, roads, and localities, as well as the dwindling availability of Andosol soils. As future expansion ensues, it presents risk of forest loss across various forest types, with pine-oak forest, mesophilic montane forest, and oyamel fir forest being of particular concern. Moreover, our results suggest that avocado production will occupy wider ranges in terms of temperature, precipitation, slope steepness and soil. The model predicts that climate change will alter the spatial distribution of avocado plantings, expanding into forest types at lower and at higher elevations. Forest loss threatens ecosystem degradation, and a wider avocado crop production footprint could lead to orchard establishment into dwindling forests that host a high diversity of native oaks and charismatic species, including the monarch butterfly.

Understanding the Geography of COVID-19 Case Fatality Rates in China: A Spatial Autoregressive Probit-Log Linear Hurdle Analysis.

This study employs a spatial autoregressive probit-log linear (SAP-Log) hurdle model to investigate the influencing factors on the probability of death and case fatality rate (CFR) of coronavirus disease 2019 (COVID-19) at the city level in China. The results demonstrate that the probability of death from COVID-19 and the CFR level are 2 different processes with different determinants. The number of confirmed cases and the number of doctors are closely associated with the death probability and CFR, and there exist differences in the CFR and its determinants between cities within Hubei Province and outside Hubei Province. The spatial probit model also presents positive spatial autocorrelation in death probabilities. It is worth noting that the medical resource sharing among cities and enjoyment of free medical treatment services of citizens makes China different from other countries. This study contributes to the growing literature on determinants of CFR with COVID-19 and has significant practical implications.

Modelling avocado-driven deforestation in Michoacán, Mexico

Abstract As demand for avocado climbs, avocado production in Michoacán - Mexico’s biggest avocado growing region - expands into new places. We use a spatial probit model to project the geographic distribution of likely future avocado expansion and analyze those results to determine (1) threats to specific forest types and (2) how the distribution of avocado is shifting spatially under current and future climate scenarios. Our results suggest that avocado expansion in Michoacán is strongly driven by distance to existing agriculture, roads, and localities, as well as the dwindling availability of Andosol soils. As future expansion ensues, it presents risk of forest loss across various forest types, with pine-oak forest, mesophilic montane forest, and oyamel fir forest being of particular concern. Moreover, our results suggest that avocado production will occupy wider ranges in terms of temperature, precipitation, slope steepness and soil. The model predicts that climate change will alter the spatial distribution of avocado plantings, expanding into forest types at lower and at higher elevations. Forest loss threatens ecosystem degradation, and a wider avocado crop production footprint could lead to orchard establishment into dwindling forests that host a high diversity of native oaks and charismatic species, including the monarch butterfly.

One or Two-Step? Evaluating GMM Efficiency for Spatial Binary Probit Models

One or Two-Step? Evaluating GMM Efficiency for Spatial Binary Probit Models

Evaluating borrowers' default risk with a spatial probit model reflecting the distance in their relational network.

Potential relationship among loan applicants can provide valuable information for evaluating default risk. However, most of the existing credit scoring models either ignore this relationship or consider a simple connection information. This study assesses the applicants’ relation in terms of their distance estimated based on their characteristics. This information is then utilized in a proposed spatial probit model to reflect the different degree of borrowers’ relation on the default prediction of loan applicant. We apply this method to peer-to-peer Lending Club Loan data. Empirical results show that the consideration of information on the spatial autocorrelation among loan applicants can provide high predictive power for defaults.

Human Rights Violations in Space: Assessing the External Validity of Machine-Geocoded versus Human-Geocoded Data

AbstractPolitical event data are widely used in studies of political violence. Recent years have seen notable advances in the automated coding of political event data from international news sources. Yet, the validity of machine-coded event data remains disputed, especially in the context of event geolocation. We analyze the frequencies of human- and machine-geocoded event data agreement in relation to an independent (ground truth) source. The events are human rights violations in Colombia. We perform our evaluation for a key, 8-year period of the Colombian conflict and in three 2-year subperiods as well as for a selected set of (non)journalistically remote municipalities. As a complement to this analysis, we estimate spatial probit models based on the three datasets. These models assume Gaussian Markov Random Field error processes; they are constructed using a stochastic partial differential equation and estimated with integrated nested Laplacian approximation. The estimated models tell us whether the three datasets produce comparable predictions, underreport events in relation to the same covariates, and have similar patterns of prediction error. Together the two analyses show that, for this subnational conflict, the machine- and human-geocoded datasets are comparable in terms of external validity but, according to the geostatistical models, produce prediction errors that differ in important respects.

Container shipping line port choice patterns in East Asia the effects of port affiliation and spatial dependence

In the container shipping sector, many shipping lines are investing in ports and terminals. Although the literature generally agrees that shipping lines can benefit financially from such vertical investments, there are conflicting views on the impacts on different stakeholders, with limited empirical evidence nor solid data analysis. In addition to being an important managerial decision for shipping lines and their affiliated ports, this may become a policy issue because public infrastructure is expected to provide equal and fair access to all consumers. This study aims to enhance the understanding of the implications of shipping line-port integration by investigating its effects on the port call choices of shipping lines at the route level. Using data from East Asia, our correlation matrix of port-calls reveals significant spatial dependence in the port call patterns of shipping lines, and complex relationships between ports in a region, which can be substitutable, complementary or no significant relationship. Our spatial probit model estimation further suggests that carriers, whether based in East Asia or elsewhere, prefer to call at affiliated ports that they own either directly or indirectly via their alliance partners. Carriers tend to avoid calling at some ports with significantly overlapping roles on the same route but are likely to call at multiple ports that are less than 1200 km apart. Our findings also indicate that shipping lines prefer ports with adequate infrastructure and that the number of ports called on one route is constrained by the total transport time. We recommend that port authorities and local governments evaluate vertical integration on a case-by-case basis, because vertical integration could help secure some port throughput but nevertheless limit competition. Further studies on the shipping line-port integration’s impacts on port handling efficiency and service quality would help policy formation on vertical integration in the maritime sector.

The impact of latent topic valence of online reviews on purchase intention for the accommodation industry

This study utilizes machine learning (ML) natural language processing (NLP) algorithms and statistical methods in order to measure the impact that qualitative textual reviews have on booking intentions for accommodations. Using over 400,000 online reviews from 1256 accommodations in South Korea, latent Dirichlet allocation (LDA) is used to determine the topic of the review content, convolutional neural networks (CNNs) are used to identify the valence of the reviews, and spatial probit models are used to determine the impact of the review content and valence on booking intention, while controlling for several other variables. It is found that positive reviews about an accommodation’s ambiance, value, service, front office, accessibility, surrounding neighborhood and room capacity result in significantly higher booking intentions, while negative reviews in the service, front office and surrounding neighborhood result in lower probability of booking. A number of explanatory variables also have varying effects on booking intentions.

Economic activity and distribution of an invasive species: Evidence from night-time lights satellite imagery data

This study addresses economic and ecological aspects related to aquatic invasive species occurrence and spread. The relative importance of these two factors driving the probability of occurrence of Elodea canadensis species in lakes of Sweden is quantified. Night-time lights data, a proxy for economic activity, is generated from satellite imagery and mapped to key environmental variables known to constrain the species' distribution. Economic and environmental data are then matched with the species' observational records at the lake scale. Evidence from a spatial probit model shows a robust positive relationship between economic activity and invasion of uninvaded lakes.

Spatial survival modelling of business re-opening after Katrina: Survival modelling compared to spatial probit modelling of re-opening within 3, 6 or 12 months

Zhou and Hanson; Zhou and Hanson; Zhou and Hanson ( 2015 , Nonparametric Bayesian Inference in Biostatistics, pages 215–46. Cham: Springer; 2018, Journal of the American Statistical Association, 113, 571–81; 2020, spBayesSurv: Bayesian Modeling and Analysis of Spatially Correlated Survival Data. R package version 1.1.4) and Zhou et al. (2020, Journal of Statistical Software, Articles, 92, 1–33) present methods for estimating spatial survival models using areal data. This article applies their methods to a dataset recording New Orleans business decisions to re-open after Hurricane Katrina; the data were included in LeSage et al. (2011b , Journal of the Royal Statistical Society: Series A (Statistics in Society), 174, 1007—27). In two articles ( LeSage etal., 2011a , Significance, 8, 160—63; 2011b, Journal of the Royal Statistical Society: Series A (Statistics in Society), 174, 1007—27), spatial probit models are used to model spatial dependence in this dataset, with decisions to re-open aggregated to the first 90, 180 and 360 days. We re-cast the problem as one of examining the time-to-event records in the data, right-censored as observations ceased before 175 businesses had re-opened; we omit businesses already re-opened when observations began on Day 41. We are interested in checking whether the conclusions about the covariates using aspatial and spatial probit models are modified when applying survival and spatial survival models estimated using MCMC and INLA. In general, we find that the same covariates are associated with re-opening decisions in both modelling approaches. We do however find that data collected from three streets differ substantially, and that the streets are probably better handled separately or that the street effect should be included explicitly.

Do accessibility and clustering affect active travel behavior in Salt Lake City?

While researchers recognize the relationship between general accessibility and active travel, it is rare to study the impact of accessibility to multi-use paths by non-motorized modes. Furthermore, auto-culture-derived issues in the U.S. imply the importance of promoting walking and cycling culture through understanding the clustering effect. Using 2012 Utah Travel Survey, this study is one of the rare that considers all typical categories of explanatory variables at two levels. First, a spatial probit model is estimated to identify whether and why people in Salt Lake City walked or cycled; second, for people who walked or cycled, a spatial autoregressive model is estimated to explore travel time by walking or cycling. The results not only reveal how accessibility to multi-use paths and attitudes influence active travel together but also report opposite spatial autocorrelations at two levels. These findings will help decision-makers efficiently decide strategical planning components to promote active transportation.

A Spatial Analysis of District Participation in Public School Choice

ABSTRACT This study examines the supply-side decision to participate in interdistrict school choice by incorporating a district’s own characteristics, those of its neighbors and a strategic component based on neighboring districts’ decisions. A spatial Durbin probit model is applied to data from Massachusetts and estimates of direct, indirect, and total effects are found. Results indicate that local capacity and average test scores affect a district’s own decision to participate. Furthermore, local socioeconomic characteristics, test scores and a district’s own decision influences neighboring district participation. Policy implications include consideration of increases in interdistrict tuition and changes in state aid for economically disadvantaged choice students.