Evaluating Apple Inc Mobility Trend Data Related to the COVID-19 Outbreak in Japan: Statistical Analysis

Background In Japan, as a countermeasure against the COVID-19 outbreak, both the national and local governments issued voluntary restrictions against going out from residences at the end of March 2020 in preference to the lockdowns instituted in European and North American countries. The effect of such measures can be studied with mobility data, such as data which is generated by counting the number of requests made to Apple Maps for directions in select countries/regions, sub-regions, and cities. Objective We investigate the associations of mobility data provided by Apple Inc and an estimate an an effective reproduction number R(t). Methods We regressed R(t) on a polynomial function of daily Apple data, estimated using the whole period, and analyzed subperiods delimited by March 10, 2020. Results In the estimation results, R(t) was 1.72 when voluntary restrictions against going out ceased and mobility reverted to a normal level. However, the critical level of reducing R(t) to <1 was obtained at 89.3% of normal mobility. Conclusions We demonstrated that Apple mobility data are useful for short-term prediction of R(t). The results indicate that the number of trips should decrease by 10% until herd immunity is achieved and that higher voluntary restrictions against going out might not be necessary for avoiding a re-emergence of the outbreak.

Download Full-text

Evaluating Apple Inc. mobility trend data related to the COVID-19 outbreak in Japan (Preprint)

10.2196/preprints.20335 ◽

2020 ◽

Author(s):

Junko Kurita ◽

Yoshiyuki Sugishita ◽

Tamie Sugawara ◽

Yasushi Ohkusa

Keyword(s):

Normal Level ◽

North American ◽

Local Governments ◽

Critical Level ◽

Polynomial Function ◽

Reproduction Number ◽

Herd Immunity ◽

Complete Cessation ◽

Normal Number ◽

Trend Data

BACKGROUND In Japan, as a countermeasure against the COVID-19 outbreak, voluntary restrictions against going out (VRG) from a residence were announced from the end of March by national and local governments in preference to lockdowns like those instituted in European and North American countries. OBJECTIVE We examined some associations among going out information provided by Apple Inc. and estimated an effective reproduction number R(t). METHODS We regressed R(t) on a polynomial function of daily Apple Inc. data. From estimation using the whole period, the sub-periods delimited by March 10 were analyzed. RESULTS Estimation results indicate R(t) as 1.72 if VRG ceases and mobility reverts to a normal level. However, the critical level of R(t)<1 was achieved at a 89.3% of the normal level of mobility. CONCLUSIONS Results indicate that a 10% reduction from the normal number of trips will be necessary until herd immunity is achieved. Complete cessation of VRG might not be necessary to avoid re-emergence of the outbreak.

Download Full-text

Prediction of the effective reproduction number of COVID-19 in Greece. A machine learning approach using Google mobility data.

10.1101/2021.05.14.21257209 ◽

2021 ◽

Author(s):

Athanasios Arvanitis ◽

Irini Furxhi ◽

Thomas Tasioulis ◽

Konstantinos Karatzas

Keyword(s):

Machine Learning ◽

Reproduction Number ◽

Short Term ◽

Attribute Importance ◽

Mobility Data ◽

Machine Learning Approach ◽

Health Related ◽

Importance Analysis ◽

Short Term Prediction ◽

Of Greece

Download Full-text

Short-term prediction of the aggregated power output of wind farms—a statistical analysis of the reduction of the prediction error by spatial smoothing effects

Journal of Wind Engineering and Industrial Aerodynamics ◽

10.1016/s0167-6105(01)00222-7 ◽

2002 ◽

Vol 90 (3) ◽

pp. 231-246 ◽

Cited By ~ 181

Author(s):

Ulrich Focken ◽

Matthias Lange ◽

Kai Mönnich ◽

Hans-Peter Waldl ◽

Hans Georg Beyer ◽

...

Keyword(s):

Statistical Analysis ◽

Power Output ◽

Prediction Error ◽

Wind Farms ◽

Spatial Smoothing ◽

Short Term ◽

Term Prediction ◽

Smoothing Effects ◽

Short Term Prediction

Download Full-text

COVID-19 immunization threshold(s): an analysis

10.1101/2021.01.02.20248596 ◽

2021 ◽

Author(s):

Luis Alfredo Bautista Balbás ◽

Mario Gil Conesa ◽

Blanca Bautista Balbás ◽

Ainhoa Alcaide Jiménez ◽

Gil Rodríguez Caravaca

Keyword(s):

Reproduction Number ◽

Herd Immunity ◽

Significant Proportion ◽

Health Programs ◽

Vaccine Research ◽

Short Term ◽

Complex Issue ◽

Public Health Programs ◽

Similar Equation ◽

Pharmaceutical Interventions

2AbstractAs COVID-19 vaccine research efforts seem to be yielding the first tangible results, the proportion of individuals needed to reap the benefits of herd immunity is a key element from a Public Health programs perspective.This magnitude, termed the critical immunization threshold (q), can be obtained from the classical SIR model equilibrium equation, equaling (1 − 1/R0)/ ϵ, where R0 is the basic reproduction number and ϵ is the vaccine efficacy. When a significant proportion of the population is already immune, this becomes (n − 1/R0)/ ϵ, where n is the proportion of non-immune individuals. A similar equation can be obtained for short-term immunization thresholds(qt), which are dependent on Rt.qs for most countries are between 60-75% of the population. Current qt for most countries are between 20-40%.Therefore, the combination of gradual vaccination and other non-pharmaceutical interventions will mark the transition to the herd immunity, providing that the later turns out to be a feasible objective. Nevertheless, immunization through vaccination is a complex issue and many challenges might appear.

Download Full-text

Mobility data can explain the entire COVID-19 outbreak course in Japan

10.1101/2020.04.26.20081315 ◽

2020 ◽

Cited By ~ 2

Author(s):

Junko Kurita ◽

Tamie Sugawara ◽

Yasushi Ohkusa

Keyword(s):

Confidence Interval ◽

Polynomial Function ◽

Reproduction Number ◽

Sir Model ◽

Control Measures ◽

Recovery Model ◽

Mobility Data

AbstractBackgroundIn Japan, as a countermeasure against the COVID-19 outbreak, voluntary restrictions against going out (VRG) have been applied.ObjectWe examined mobility information provided by Apple Inc. to a susceptible–infected–recovery model.MethodWhen applying a polynomial function to daily Apple data with the SIR model, we presumed the function up to a cubic term as in our earlier study.ResultsEstimation results demonstrated R0 as 1.507 and its 95% confidence interval was [1.502, 1.509].. The estimated coefficients of Apple data was 1.748 and its 95% confidence interval was [1.731, 1.788].Discussion and ConclusionResults show that mobility data from Apple Inc. can explain the entire course of the outbreak in COVID-19 in Japan. Therefore, monitoring Apple data might be sufficient to adjust control measures to maintain an effective reproduction number of less than one.

Download Full-text

Prediction of the effective reproduction number of COVID-19 in Greece. A machine learning approach using Google mobility data

Journal of Decision Analytics and Intelligent Computing ◽

10.31181/jdaic1001202201f ◽

2021 ◽

Vol 1 (1) ◽

pp. 1-21

Author(s):

Athanasios Arvanitis ◽

◽

Irini Furxhi ◽

Thomas Tasioulis ◽

Konstantinos Karatzas ◽

...

Keyword(s):

Machine Learning ◽

Reproduction Number ◽

The Other ◽

Attribute Importance ◽

Mobility Data ◽

Machine Learning Approach ◽

Health Related ◽

Importance Analysis ◽

Short Term Prediction ◽

Of Greece

This paper demonstrates how a short-term prediction of the effective reproduction number (Rt) of COVID-19 in regions of Greece is achieved based on online mobility data. Various machine learning methods are applied to predict Rt and attribute importance analysis is performed to reveal the most important variables that affect the accurate prediction of Rt. Work and Park categories are identified as the most important mobility features when compared to the other attributes, with values of 0.25 and 0.24, respectively. Our results are based on an ensemble of diverse Rt methodologies to provide non-precautious and non-indulgent predictions. Random Forest algorithm achieved the highest R2 (0.8 approximately), Pearson’s and Spearman’s correlation values close to 0.9, outperforming in all metrics the other models. The model demonstrates robust results and the methodology overall represents a promising approach towards COVID-19 outbreak prediction. This paper can help health-related authorities when deciding on non-nosocomial interventions to prevent the spread of COVID-19.

Download Full-text

Importance of suppression and mitigation measures in managing COVID-19 outbreaks

10.1101/2020.03.31.20048835 ◽

2020 ◽

Cited By ~ 3

Author(s):

Michael E. Hochberg

Keyword(s):

Control Strategy ◽

Reproduction Number ◽

Herd Immunity ◽

Mitigation Measures ◽

Reproductive Number ◽

Short Term ◽

Community Size ◽

Epidemic Curve ◽

Total Community ◽

Step Control

AbstractI employ a simple mathematical model of an epidemic process to evaluate how four basic quantities: the reproduction number , the numbers of sensitive (S) and infectious individuals (I), and total community size (N) affect strategies to control COVID-19. Numerical simulations show that strict suppression measures at the beginning of an epidemic can create low infectious numbers, which thereafter can be managed by mitigation measures over longer periods to flatten the epidemic curve. The stronger the suppression measure, the faster it achieves the low numbers of infections that are conducive to subsequent management. Our results on short-term strategies point to either a two-step control strategy, following failed mitigation, that begins with suppression of the reproduction number, , below 1.0, followed by renewed mitigation measures that manage the epidemic by maintaining at approximately 1.0, or should suppression not be feasible, the progressive lowering of the effective reproductive number, , below 1.0. The objectives of the full sequence of measures observed in a number of countries, and likely to see in the longer term, can be symbolically represented as: . We discuss the predictions of this analysis and how it fits into longer-term sequences of measures, including misconceptions about ‘flattening the curve’ and how the herd immunity concept can be used to ‘leverage’ immunity.

Download Full-text