IDEAS home Printed from https://ideas.repec.org/p/ags/aaea16/236124.html
   My bibliography  Save this paper

Forecasting Agri-food Consumption Using the Keyword Volume Index from Search Engine Data

Author

Listed:
  • Ikhoon, Jang
  • Young Chan, Choe

Abstract

Forecasting in the agri-food sector is an important topic. Accurate yield prediction improves farmers’ revenue and management stability. Agile demand forecasting helps agri-food producers to adjust their production properly. Also, appropriate price forecasting directly affect the profit of agri-food suppliers and distributors, and can be important information for decision making by policy makers. However, previous studies often use only a price variable as a predictor without critical variables such as yield or demand because collecting instant yield or demand data is difficult. Recently, with the progress of technology, performance of production yield forecasting using climate data have reached a high level. For example, Monsanto acquires Climate Corporation, a climate data research company for $930 million to maximize farmers’ yields. This event means that yield forecasting techniques have been matured.However, studies on forecasting agri-food demands are not vibrant due to lack of real-time consumer data. Presently, there are several sources of data on real-time economic activity available from private sector companies such as Google, MasterCard, Federal Express, UPS, and Intuit (Choi & Varian, 2012). Actually, many forecasting studies with significant results have used search engine data as an important indicator to improve prediction accuracy. Yahoo’s search query predicted stock market trade volume (Preis et al., 2013). Google Trends data forecasted the number of movie viewers (Hand & Judge, 2011), which is used as an important predictor and indicator of monthly consumption (Vosen & Schmidt, 2011; Kholodilin et al., 2010). In the commodity markets, automobile purchases (Choi & Varian, 2012; Carrière-Swallow & Labbé, 2013) and house prices (Wu & Brynjolfsson, 2014) can be explained by Google Trends data. However, there are few studies on forecasting using Google Trends data in the agri-food and agribusiness sectors. In this study, we examine Naver Trend of South Korea (similar to Google Trends), which is a real-time weekly index of the volume of queries that users enter into Naver. We aim to find that these search engine data improves accuracy in forecasting consumption of agri-foods. For a more detail explanation, we classify empirically agri-food items into several specific groups. Consumption data is collected from a panel survey of housewives from the Rural Development Administration in South Korea. The data includes five years of daily household food consumption records based on receipt from December 2009 through November 2014. We finally obtained 3.6 million purchase data of 732 panels continuously sustained during five years. The data set for estimation is processed on a weekly based amount of consumption by targeted agri-food items for 261 weeks. The prediction model of this study is based on the AR(1) and seasonal AR(1) model of Choi and Varian (2012). The baseline model only consists of the consumption of t period and lagged consumption of t-1 period. The AR(1) term (in the case of seasonal AR(1) and lagged consumption of t-52 period) and the test model have an additional predictor as a keyword volume index of t period. A 3-week moving average is used for the model estimation in order to smooth the short-term fluctuation of the measured value. For the evaluation of prediction models, we first examine that the test model is a significantly improved in-sample fit, and we next investigate whether the trends variables improve out-of-sample forecasting by the rolling window method, which is consistent with the method of Choi and Varian (2012). The model estimation and prediction performance evaluation is achieved by the agri-food items when considering the classified groups. The first classification measure is whether the agri-food item is a main (or single) dish or minor ingredients of a main dish. The second measure is whether the agri-food item is consumed at home or for eating out. The third measure is whether the agri-food item is a common product or a brand product. The fourth measure is whether the agri-food item has macro sales trends or no trends. The results are different by agri-food groups. The agri-food items often used for main or single dishes such as chicken, sweet pumpkin, sweet potato, and shiitake mushroom tend to show significant improvement for both in-sample estimation and out-of-sample forecast. However, agri-food items used mainly as minor ingredients such as carrot, cucumber, canned tuna, and enoki mushroom are not significant. Even if the item is a main or single dish, agri-food items frequently consumed out of the home such as roasted pork belly, fried chicken, and dumpling have no significant relationship. However, agri-food items mainly consumed at home such as most fruits are significant. Meanwhile, the significant difference between common products and brand products are not found. Rather, the more important criteria are macro trends. Agri-food items with gradual growth macro trends or long-term fluctuation have a significant relationship. For example, instant rice products, healthy food such as chicken breast and nut products, or newly released ramen products with a growing trend show a significant relationship. On the contrary, the steady selling of ramen products or dumpling products is not significant. Although the Naver Trends data for all of the agri-food items do not improve performance of consumption forecasting, the search engine data is correlated with actual consumption in several agri-food groups such as 1)the ingredients of main or single dishes, and 2)ingredients mainly consumed at home, and 3)products with macro sales trends. The first and second groups are directly linked with the searching activities for actual agri-food consumption. The third group has volatility for the estimation in the difference of steady selling products with low fluctuation. Therefore, search engine data of the agri-foods with these characteristics can be considered as an important predictor of agri-food consumption forecasting research. This study shows that consumers’ internet searching activity-related data rarely dealt with previous research in the agri-food sector, which can be a meaningful factor to predict more accurate actual consumption. We expect to extend many further relevant studies, and to help agri-food producers and distributors when they forecast the sales of their agri-products

Suggested Citation

  • Ikhoon, Jang & Young Chan, Choe, 2016. "Forecasting Agri-food Consumption Using the Keyword Volume Index from Search Engine Data," 2016 Annual Meeting, July 31-August 2, Boston, Massachusetts 236124, Agricultural and Applied Economics Association.
  • Handle: RePEc:ags:aaea16:236124
    DOI: 10.22004/ag.econ.236124
    as

    Download full text from publisher

    File URL: https://ageconsearch.umn.edu/record/236124/files/AAEA2016_full_Jang_160525.pdf
    Download Restriction: no

    File URL: https://libkey.io/10.22004/ag.econ.236124?utm_source=ideas
    LibKey link: if access is restricted and if your library uses this service, LibKey will redirect you to where you can use your library subscription to access this item
    ---><---

    References listed on IDEAS

    as
    1. Konstantin A. Kholodilin & Maximilian Podstawski & Boriss Siliverstovs, 2010. "Do Google Searches Help in Nowcasting Private Consumption?: A Real-Time Evidence for the US," Discussion Papers of DIW Berlin 997, DIW Berlin, German Institute for Economic Research.
    2. Konstantin Kholodilin & Maximilian Podstawski & Boriss Siliverstovs, 2010. "Do Google Searches Help in Nowcasting Private Consumption?," KOF Working papers 10-256, KOF Swiss Economic Institute, ETH Zurich.
    3. Hyunyoung Choi & Hal Varian, 2012. "Predicting the Present with Google Trends," The Economic Record, The Economic Society of Australia, vol. 88(s1), pages 2-9, June.
    4. Guney, Selin, 2015. "An evaluation of price forecasts of the cattle market under structural changes," 2015 AAEA & WAEA Joint Annual Meeting, July 26-28, San Francisco, California 205109, Agricultural and Applied Economics Association.
    5. Shang, Xia & Tonsor, Glynn T., 2015. "Comparing Forecasting Ability of Demand System Using Different Data Sources: the Case of U.S. Meat Demand with Food Safety Recalls," 2015 AAEA & WAEA Joint Annual Meeting, July 26-28, San Francisco, California 205624, Agricultural and Applied Economics Association.
    6. Chris Hand & Guy Judge, 2012. "Searching for the picture: forecasting UK cinema admissions using Google Trends data," Applied Economics Letters, Taylor & Francis Journals, vol. 19(11), pages 1051-1055, July.
    7. Xu, Xiaojie & Thurman, Walter, 2015. "Forecasting Local Grain Prices: An Evaluation of Composite Models in 500 Corn Cash Markets," 2015 AAEA & WAEA Joint Annual Meeting, July 26-28, San Francisco, California 205332, Agricultural and Applied Economics Association.
    8. Allen, P. Geoffrey, 1994. "Economic forecasting in agriculture," International Journal of Forecasting, Elsevier, vol. 10(1), pages 81-135, June.
    Full references (including those not matched with items on IDEAS)

    Most related items

    These are the items that most often cite the same works as this one and are cited by the same works as this one.
    1. Palma Lampreia Dos Santos, Maria José, 2018. "Nowcasting and forecasting aquaponics by Google Trends in European countries," Technological Forecasting and Social Change, Elsevier, vol. 134(C), pages 178-185.
    2. Chien-jung Ting & Yi-Long Hsiao, 2022. "Nowcasting the GDP in Taiwan and the Real-Time Tourism Data," Advances in Management and Applied Economics, SCIENPRESS Ltd, vol. 12(3), pages 1-2.
    3. Nikolaos Askitas & Klaus F. Zimmermann, 2015. "The internet as a data source for advancement in social sciences," International Journal of Manpower, Emerald Group Publishing Limited, vol. 36(1), pages 2-12, April.
    4. Jianchun Fang & Wanshan Wu & Zhou Lu & Eunho Cho, 2019. "Using Baidu Index To Nowcast Mobile Phone Sales In China," The Singapore Economic Review (SER), World Scientific Publishing Co. Pte. Ltd., vol. 64(01), pages 83-96, March.
    5. Jingwen Liu & Peng Zou & Yu Ma, 2022. "The Effect of Air Pollution on Food Preferences," Journal of the Academy of Marketing Science, Springer, vol. 50(2), pages 410-423, March.
    6. Chien-jung Ting & Yi-Long Hsiao & Rui-jun Su, 2022. "Application of the Real-Time Tourism Data in Nowcasting the Service Consumption in Taiwan," Journal of Applied Finance & Banking, SCIENPRESS Ltd, vol. 12(4), pages 1-4.
    7. Siliverstovs, Boriss & Wochner, Daniel S., 2018. "Google Trends and reality: Do the proportions match?," Journal of Economic Behavior & Organization, Elsevier, vol. 145(C), pages 1-23.
    8. David Iselin & Boriss Siliverstovs, 2013. "Using Newspapers for Tracking the Business Cycle," KOF Working papers 13-337, KOF Swiss Economic Institute, ETH Zurich.
    9. David Iselin & Boriss Siliverstovs, 2013. "Mit Zeitungen Konjunkturprognosen erstellen: Eine Vergleichsstudie für die Schweiz und Deutschland," KOF Analysen, KOF Swiss Economic Institute, ETH Zurich, vol. 7(3), pages 104-117, September.
    10. Gomes, Pedro & Taamouti, Abderrahim, 2016. "In search of the determinants of European asset market comovements," International Review of Economics & Finance, Elsevier, vol. 44(C), pages 103-117.
    11. Nuno Barreira & Pedro Godinho & Paulo Melo, 2013. "Nowcasting unemployment rate and new car sales in south-western Europe with Google Trends," Netnomics, Springer, vol. 14(3), pages 129-165, November.
    12. Tong Liu & Guojun He & Alexis Lau, 2018. "Avoidance behavior against air pollution: evidence from online search indices for anti-PM2.5 masks and air filters in Chinese cities," Environmental Economics and Policy Studies, Springer;Society for Environmental Economics and Policy Studies - SEEPS, vol. 20(2), pages 325-363, April.
    13. Karaman Örsal, Deniz Dilan, 2021. "Onlinedaten und Konsumentscheidungen: Voraussagen anhand von Daten aus Social Media und Suchmaschinen," Edition HWWI: Chapters, in: Straubhaar, Thomas (ed.), Neuvermessung der Datenökonomie, volume 6, pages 157-172, Hamburg Institute of International Economics (HWWI).
    14. Yang, Xin & Pan, Bing & Evans, James A. & Lv, Benfu, 2015. "Forecasting Chinese tourist volume with search engine data," Tourism Management, Elsevier, vol. 46(C), pages 386-397.
    15. Dirk Ulbricht & Konstantin A. Kholodilin & Tobias Thomas, 2017. "Do Media Data Help to Predict German Industrial Production?," Journal of Forecasting, John Wiley & Sons, Ltd., vol. 36(5), pages 483-496, August.
    16. Rose, Andrew K. & Spiegel, Mark M., 2012. "Dollar illiquidity and central bank swap arrangements during the global financial crisis," Journal of International Economics, Elsevier, vol. 88(2), pages 326-340.
    17. Perroni, Carlo & Scharf, Kimberley & Talavera, Oleksandr & Vi, Linh, 2022. "Does online salience predict charitable giving? Evidence from SMS text donations," Journal of Economic Behavior & Organization, Elsevier, vol. 197(C), pages 134-149.
    18. Fondeur, Y. & Karamé, F., 2013. "Can Google data help predict French youth unemployment?," Economic Modelling, Elsevier, vol. 30(C), pages 117-125.
    19. Abay,Kibrom A. & Hirfrfot,Kibrom Tafere & Woldemichael,Andinet, 2020. "Winners and Losers from COVID-19 : Global Evidence from Google Search," Policy Research Working Paper Series 9268, The World Bank.
    20. Stephen L. France & Yuying Shi, 2017. "Aggregating Google Trends: Multivariate Testing and Analysis," Papers 1712.03152, arXiv.org, revised Mar 2018.

    More about this item

    Keywords

    Agribusiness;

    Statistics

    Access and download statistics

    Corrections

    All material on this site has been provided by the respective publishers and authors. You can help correct errors and omissions. When requesting a correction, please mention this item's handle: RePEc:ags:aaea16:236124. See general information about how to correct material in RePEc.

    If you have authored this item and are not yet registered with RePEc, we encourage you to do it here. This allows to link your profile to this item. It also allows you to accept potential citations to this item that we are uncertain about.

    If CitEc recognized a bibliographic reference but did not link an item in RePEc to it, you can help with this form .

    If you know of missing items citing this one, you can help us creating those links by adding the relevant references in the same way as above, for each refering item. If you are a registered author of this item, you may also want to check the "citations" tab in your RePEc Author Service profile, as there may be some citations waiting for confirmation.

    For technical questions regarding this item, or to correct its authors, title, abstract, bibliographic or download information, contact: AgEcon Search (email available below). General contact details of provider: https://edirc.repec.org/data/aaeaaea.html .

    Please note that corrections may take a couple of weeks to filter through the various RePEc services.

    IDEAS is a RePEc service. RePEc uses bibliographic data supplied by the respective publishers.