The use of Twitter data to improve small area estimates of households’ share of food consumption expenditure in Italy

Keogh

J. of Data, Inf. and Manag.

2022

The emergence of big data (BD) offers new opportunities for food businesses to address emerging risks and operational challenges. BD denotes the integration and analysis of multiple data sets, which are inherently complex, voluminous and are often of inadequate quality and structure. While BD is a well-established method in supply chain management, academic research on its application in the food ecosystem is still lagging. To fill this knowledge gap and capture the latest developments in this field, a systematic literature review was performed. Forty-one papers were selected and thoroughly examined and analysed to identify the enablers of BD in the food supply chain. The review primarily attempted to obtain an answer to the following research question: “What are the possibilities of leveraging big data in the food supply chain?“ Six significant benefits of applying BD in the food industry were identified, namely, the extraction of valuable knowledge and insights, decision-making support, improvement of food chain efficiencies, reliable forecasting, waste minimization, and food safety. Finally, some challenges and future research directions were outlined.

Section: Introductionmentioning

confidence: 99%

Big data in the food supply chain: a literature review

Keogh

J. of Data, Inf. and Manag.

2022

Journal of Official Statistics

“…Here, we would contend that adopting a system of estimation that accounts for both sampling streams, yet incentivizes probability-based observations and allows for the direct quantification of uncertainty in survey estimates is a more defensible strategy than one that renounces probability sampling entirely along with all of its attractive theoretical properties. Moreover, the idea of enhancing a small, but carefully designed, probability sample with abundant but potentially error-prone data is not a new idea and is a widely accepted strategy in small area applications where sparse probability samples are routinely supplemented with alternative data sources to improve the cost and error properties of population estimates (Marchetti et al 2016;Porter et al 2014;Briggs et al 2007;Schmertmann et al 2013).…”

Section: Discussionmentioning

confidence: 99%

Supplementing Small Probability Samples with Nonprobability Samples: A Bayesian Approach

Sakshaug

Wiśniowski

Ruiz

et al. 2019

Carefully designed probability-based sample surveys can be prohibitively expensive to conduct. As such, many survey organizations have shifted away from using expensive probability samples in favor of less expensive, but possibly less accurate, nonprobability web samples. However, their lower costs and abundant availability make them a potentially useful supplement to traditional probability-based samples. We examine this notion by proposing a method of supplementing small probability samples with nonprobability samples using Bayesian inference. We consider two semi-conjugate informative prior distributions for linear regression coefficients based on nonprobability samples, one accounting for the distance between maximum likelihood coefficients derived from parallel probability and non-probability samples, and the second depending on the variability and size of the nonprobability sample. The method is evaluated in comparison with a reference prior through simulations and a real-data application involving multiple probability and nonprobability surveys fielded simultaneously using the same questionnaire. We show that the method reduces the variance and mean-squared error (MSE) of coefficient estimates and model-based predictions relative to probability-only samples. Using actual and assumed cost data we also show that the method can yield substantial cost savings (up to 55%) for a fixed MSE.

Journal of Official Statistics

“…This work is an attempt to implement their idea in a systematic way. Marchetti et al (2016) instead, used data coming from Twitter (Curini et al 2015) as an instrumental covariate to estimate the Italian household share of food consumption expenditures at a provincial level, that is, they exploit the correlation between the official statistics indicator and social media data at regional level to reconstruct the official statistics at sub-regional level, thanks to the granularity of the Twitter data.…”

Section: General Sae Modelsmentioning

confidence: 99%

Controlling for Selection Bias in Social Media Indicators through Official Statistics: a Proposal

Iacus

Porro

Salini

et al. 2020

AbstractWith the increase of social media usage, a huge new source of data has become available. Despite the enthusiasm linked to this revolution, one of the main outstanding criticisms in using these data is selection bias. Indeed, the reference population is unknown. Nevertheless, many studies show evidence that these data constitute a valuable source because they are more timely and possess higher space granularity. We propose to adjust statistics based on Twitter data by anchoring them to reliable official statistics through a weighted, space-time, small area estimation model. As a by-product, the proposed method also stabilizes the social media indicators, which is a welcome property required for official statistics. The method can be adapted anytime official statistics exists at the proper level of granularity and for which social media usage within the population is known. As an example, we adjust a subjective well-being indicator of “working conditions” in Italy, and combine it with relevant official statistics. The weights depend on broadband coverage and the Twitter rate at province level, while the analysis is performed at regional level. The resulting statistics are then compared with survey statistics on the “quality of job” at macro-economic regional level, showing evidence of similar paths.