Using Google Trends Data to forecast homicide mortality: the case of Mexico

Authors

DOI:

https://doi.org/10.15517/c57dzg81

Keywords:

Homicides, forecasting, Google Trends, VAR model

Abstract

Introduction: In Mexico a major public safety concern is how to predict and reduce homicides to implement effective mitigation policies. Methodology: This study aims to compare traditional forecasting models —ARIMA and Vector Autoregressive (VAR)—with and without Google Trends data, the research explores ways to enhance prediction accuracy. Using homicide records from the National Institute of Statistics and Geography (INEGI, for its Spanish acronym) and Google Trends data from 2006–2020, the study highlights the integration of real-time online data to complement official statistics. Results: Considering a forecast horizon of 15 months up to March 2020, results show that VAR models with Google Trends provide the best performance for both female and male homicides. Conclusions: The findings underscore the potential of integrating digital data sources into traditional models to provide more accurate and timely tools for public safety planning and intervention.

Downloads

Download data is not yet available.

References

Box, G. E., Jenkins, G. M., Reinsel, G. C., & Ljung, G. M. (2015). Time series analysis: Forecasting and control. John Wiley & Sons.

Calvo, H., Godoy-Calderón, S., Moreno-Armendáriz, M. A., & Martínez-Hernández, V. M. (2017). Forecasting, clustering and patrolling criminal activities, Intelligent Data Analysis, 21(4), 697–720. https://doi.org/10.3233/IDA-170883

Cebrián, E., & Domenech, J. (2024). Addressing Google Trends inconsistencies, Technological Forecasting and Social Change, 202, 123318. https://doi.org/10.1016/j.techfore.2024.123318

Clements, M. P., & Hendry, D. F. (Eds.). (2011). The Oxford handbook of economic forecasting. Oxford University Press. https://doi.org/10.1093/oxfordhb/9780195398649.001.0001

Delgadillo, G., & Torres, D. (2023). ¿Qué es y cómo funciona el Registro Nacional de Personas Desaparecidas y No Localizadas (RNPDNO)? [What is and how does the National Registry of Disappeared and Unlocated Persons (RNPDNO) work?]. Animal Político. https://animalpolitico.com/analisis/organizaciones/el-blog-del-seminario-sobre-violencia-y-paz/registro-nacional-personas-desaparecidas-que-es-como-funciona?rtbref=rtb_krqfxr3j12dhx0dkl7c2_1714875535898

Fabrigar, L. R., Wegener, D. T., MacCallum, R. C., & Strahan, E. J. (1999). Evaluating the use of exploratory factor analysis in psychological research. Psychological Methods, 4(3), 272–299. https://doi.org/10.1037/1082-989X.4.3.272

García-Gómez, J., Valdez, S., & Carlos, H. (2022). Homicide forecasting for the state of Guanajuato using LSTM and geospatial information in 2022 IEEE Mexican International Conference on Computer Science (ENC), 1-6. https://doi.org/10.1109/ENC56672.2022.9882957

Gujarati, D. N., & Porter, D. C. (2009). Basic econometrics. McGraw-Hill Companies.

Gutiérrez, F., & Hernández, D. M. (2024). Desaparición forzada [Forced disappearance]. Constructos Criminológicos, 4(6), 43–56. https://doi.org/10.29105/cc4.6-59

Hernández-Gress, E., Flegl, M., Krstikj, A., & Boyes, C. (2023). Femicide in Mexico: statistical evidence of an increasing trend, PLOS ONE, 18(12), e0290165. https://doi.org/10.1371/journal.pone.0290165

Instituto Nacional de Estadística y Geografía. (1990-2023). Estadísticas vitales de Mortalidad, 1990 – 2023. Dirección General de Estadística, México.

Hyvärinen, A., & Oja, E. (2000). Independent component analysis: Algorithms and applications. Neural Networks, 13(4–5), 411–430. https://doi.org/10.1016/S0893-6080(00)00026-5

Jolliffe, I. (2013). Principal component analysis, Springer Science & Business Media.

Massicotte, P., & Eddelbuettel, D. (2025). gtrendsR: Perform and display Google Trends queries (Version 1.5.2) [R package]. Comprehensive R Archive Network (CRAN). https://cran.r-project.org/package=gtrendsR

Medeiros, M. C., & Pires, H. F. (2021). The proper use of Google Trends in forecasting models, arXiv:2104.03065. https://doi.org/10.48550/arXiv.2104.03065

Observatorio Nacional Ciudadano. (2017). Desapariciones forzadas e involuntarias. El Registro Estadístico de la Desaparición: ¿Delito o Circunstancia? [Forced and involuntary disappearances. The Statistical Registry of Disappearance: Crime or Circumstance?]. Ciudad de México.

Pearsall, B. (2010). Predictive policing: The future of law enforcement, National Institute of Justice Journal, 266(1), 16-19.

Piña-García, C. A., & Ramírez-Ramírez, L. (2019). Exploring crime patterns in Mexico City, Journal of Big Data, 6, 1-21. https://doi.org/10.1186/s40537-019-0228-x

Ramallo, S., Camacho, M., Ruiz Marín, M., & Porfiri, M. (2023). A dynamic factor model to predict homicides with firearms in the United States, Journal of Criminal Justice, 86, 102051. https://doi.org/10.1016/j.jcrimjus.2023.102051

Sahni, S. P., & Phakey, N. (2021). Criminal psychology: Understanding criminal behaviour, Criminal Psychology and the Criminal Justice System in India and Beyond, 21-30. https://doi.org/10.1007/978-981-16-4570-9_2

Santos‐Marquez, F. (2021). Spatial beta‐convergence forecasting models: Evidence from municipal homicide rates in Colombia, Journal of Forecasting, 41(2), 294–302. https://doi.org/10.1002/for.2816

Sims, C. A. (1980). Macroeconomics and Reality, Econometrica, 48(1), 1–48. https://www.jstor.org/stable/1912017

Swedo, E. A., Alic, A., Law, R. K., Sumner, S. A., Chen, M. S., Zwald, M. L., Van Dyke, M. E., Bowen, D. A., & Mercy, J. A. (2023). Development of a machine learning model to estimate US firearm homicides in near real time, JAMA Network Open, 6(3), e233413. https://doi.org/10.1001/jamanetworkopen.2023.3413

United Nations Office on Drugs and Crime. (2019). Global study on homicide. https://www.unodc.org/documents/data-and-analysis/gsh/Booklet1.pdf

Wei, W. W. (2019). Multivariate time series analysis and applications, John Wiley & Sons.

World Health Organization. (2015). International Statistical Classification of Diseases and Related Health Problems: 10th Revision (ICD-10) (5th 2016 Revision ed.). World Health Organization.

Downloads

Published

2025-09-24

Issue

Volume 23, Issue 1: july-december 2024 (in progress)

Section

Scientific articles

License

Copyright (c) 2025 Vazquez & Silva

Creative Commons License

This work is licensed under a Creative Commons Attribution-NonCommercial-ShareAlike 4.0 International License.

Population and Health in Mesoamerica informs its authors and readers that all publications must be submitted with the Letter of Acceptance of Copyright Conditions where the authors are asked for authorization to submit the article to the Journal for consideration so that it may carry out any of the following activities:

  1. The graphic and style editing of the work or part of it.
    The publication and reproduction of the work or part of it, both in print and electronic media, including the Internet and any other known or unknown technology.
    The translation of the work or part of it into any language or dialect.
    The adaptation of the work to reading, sound, voice and any other available technical representation or mechanism, which makes possible its access for partially or totally blind people, or with any other form of special abilities that prevent their access to the conventional reading of the article.
    The distribution and availability of the work to the public, in such a way that the public may have access to them from the time and place that each person chooses, through the physical or electronic mechanisms available to them.
    Any other form of use, process or system known or to be known that is related to the activities and editorial purposes to which the Journal is linked.

You can access the Letter by clicking HERE. If you have a question, please write to revista.ccp@ucr.ac.cr

In addition, the content of this site is protected under the Attribution - Noncommercial - Share Alike (by-nc-sa) license which permits commercial use of the original work or derivative works and distribution of which must be made under the same license that governs the original work. The graphic design, images and texts generated by the e-Information Science magazine are property of the Universidad de Costa Rica and are licensed under Creative Commons Attribution - NonCommercial - Attribution-ShareAlike 4.0 International.

Authors who publish in this journal accept the following conditions:

-The authors retain their moral rights over the publication and assign the patrimonial rights mentioned in the Letter of Assignment of Rights with the Creative Commons Attribution - NonCommercial - Compatitr Equal 4.0 International license, which allows third parties to use what is published as long as they mention the authorship of the work and to the first publication in this journal.

-Authors may make other independent and additional contractual arrangements for non-exclusive distribution of the version of the article published in this journal (e.g., inclusion in an institutional repository or publication in a book) as long as they clearly indicate that the work was first published in this journal.

This policy is established in accordance with Law 6638 on Copyright and Related Rights of the Republic of Costa Rica.